52 results
Search Results
2. Build TensorFlow Input Pipelines
- Author
-
David Paper
- Subjects
Pipeline transport ,Engineering drawing ,Sequence ,Computer science ,Simple (abstract algebra) ,business.industry ,Component (UML) ,Deep learning ,Artificial intelligence ,Learning models ,business ,Abstraction (linguistics) - Abstract
We introduce you to TensorFlow input pipelines with the tf.data API, which enables you to build complex input pipelines from simple, reusable pieces. Input pipelines are the lifeblood of any deep learning experiment because learning models expect data in a TensorFlow consumable form. It is very easy to create high-performance pipelines with the tf.data.Dataset abstraction (a component of the tf.data API) because it represents a sequence of elements from a dataset in a simple format.
- Published
- 2021
3. Deep Learning with TensorFlow Datasets
- Author
-
David Paper
- Subjects
business.industry ,Computer science ,Simple (abstract algebra) ,Deep learning ,Artificial intelligence ,Machine learning ,computer.software_genre ,business ,computer - Abstract
In the previous chapter, we demonstrated how to work with TFDS objects. In this chapter, we work through two end-to-end deep learning experiments with large and complex TFDS objects. The Fashion-MNIST and beans datasets are small with simple images.
- Published
- 2021
4. Automated Text Generation
- Author
-
David Paper
- Subjects
Network architecture ,Ideal (set theory) ,Artificial neural network ,Contextual image classification ,business.industry ,Computer science ,Feed forward ,Pattern recognition ,ComputingMethodologies_PATTERNRECOGNITION ,Flow (mathematics) ,Text generation ,Artificial intelligence ,Layer (object-oriented design) ,business - Abstract
Feedforward neural nets are generally great for classification and regression problems. CNNs are great for complex image classification. But activations for feedforward nets and CNNs flow only in one direction, from the input layers to the output layer. Since signals flow in only one direction, feedforward and convolutional nets are not ideal if patterns in data change over time. So we need a different network architecture to work with data impacted by time.
- Published
- 2021
5. Time Series Forecasting with RNNs
- Author
-
David Paper
- Subjects
Multivariate statistics ,Computer science ,business.industry ,Univariate ,Artificial intelligence ,Time series ,business ,Machine learning ,computer.software_genre ,computer - Abstract
We’ve already leveraged RNNs for NLP. In this chapter, we create experiments to forecast with time series data. We use the famous Weather dataset to demonstrate both a univariate and a multivariate example.
- Published
- 2021
6. Increase the Diversity of Your Dataset with Data Augmentation
- Author
-
David Paper
- Subjects
Training set ,business.industry ,Computer science ,Deep learning ,Artificial intelligence ,business ,Machine learning ,computer.software_genre ,computer ,Diversity (business) - Abstract
We guide you in the creation of augmented data experiments to increase the diversity of a training set by applying random (but realistic) transformations. Data augmentation is very useful for small datasets because deep learning models crave a lot of data to perform well.
- Published
- 2021
7. An Introduction to Reinforcement Learning
- Author
-
David Paper
- Subjects
Intelligent agent ,business.industry ,Computer science ,Order (business) ,media_common.quotation_subject ,Reinforcement learning ,Artificial intelligence ,business ,computer.software_genre ,Function (engineering) ,computer ,ComputingMilieux_MISCELLANEOUS ,media_common - Abstract
Reinforcement learning (RL) is an area of machine learning that focuses on teaching intelligent agents how to take actions in an environment in order to maximize cumulative reward. Cumulative reward in RL is the sum of all rewards as a function of the number of training steps.
- Published
- 2021
8. Convolutional and Variational Autoencoders
- Author
-
David Paper
- Subjects
business.industry ,Computer science ,Feed forward ,Pattern recognition ,Artificial intelligence ,business - Abstract
Autoencoders don’t typically work well with images unless they are very small. But convolutional and variational autoencoders work much better than feedforward dense ones with large color images.
- Published
- 2021
9. Convolutional Neural Networks
- Author
-
David Paper
- Subjects
Computer science ,business.industry ,Computer Science::Neural and Evolutionary Computation ,Pattern recognition ,Space (mathematics) ,Convolutional neural network ,ComputingMethodologies_PATTERNRECOGNITION ,Simple (abstract algebra) ,Computer Science::Computer Vision and Pattern Recognition ,Feedforward neural network ,Pixel matrix ,Artificial intelligence ,business ,Training performance ,MNIST database - Abstract
With feedforward neural networks, we achieved good training performance with MNIST and Fashion-MNIST datasets. But images in these datasets are simple and centered within the input space that contains them. That is, they are centered within the pixel matrix that holds them. Input space is all the possible inputs to a model.
- Published
- 2021
10. Build Your First Neural Network with Google Colab
- Author
-
David Paper
- Subjects
World Wide Web ,Work (electrical) ,Artificial neural network ,business.industry ,Computer science ,Deep learning ,Cloud computing ,Artificial intelligence ,Python (programming language) ,business ,computer ,computer.programming_language - Abstract
We work through a complete deep learning example with Python’s TensorFlow 2.x library in the Google Colab cloud service. We also demonstrate how to link your Google Drive with the Colab cloud service.
- Published
- 2021
11. Simple Transfer Learning with TensorFlow Hub
- Author
-
David Paper
- Subjects
Artificial neural network ,Download ,business.industry ,Process (engineering) ,Computer science ,Learning models ,Open source ,Scratch ,Simple (abstract algebra) ,Artificial intelligence ,business ,Transfer of learning ,computer ,computer.programming_language - Abstract
Transfer learning is the process of creating new learning models by fine-tuning previously trained neural networks. Instead of training a network from scratch, we download a pre-trained open source learning model and fine-tune it for our own purpose. A pre-trained model is one that is created by someone else to solve a similar problem. We can use one of these instead of building our own model. A big advantage is that a pre-trained model has been crafted by experts, so we can be confident that it performs at a high level (in most cases). Another advantage is that we don’t have to have a lot of data to use a pre-trained model.
- Published
- 2021
12. Generative Adversarial Networks
- Author
-
David Paper
- Subjects
Adversarial system ,Generative model ,Training set ,business.industry ,Computer science ,Unsupervised learning ,Artificial intelligence ,business ,Generative adversarial network ,Generative grammar ,Generative modeling - Abstract
Generative modeling is an unsupervised learning technique that involves automatically discovering and learning the regularities (or patterns) in input data so that a trained model can generate new examples that plausibly could have been drawn from the original dataset. A popular type of generative model is a generative adversarial network. Generative adversarial networks (GANs) are generative models that create new data instances that resemble the training data.
- Published
- 2021
13. Introduction to Tensor Processing Units
- Author
-
David Paper
- Subjects
Tensor processing unit ,business.industry ,Computer science ,Deep learning ,Integrated circuit ,law.invention ,Computer engineering ,Application-specific integrated circuit ,law ,Tensor (intrinsic definition) ,Code (cryptography) ,Google Brain ,Artificial intelligence ,business - Abstract
We introduce you to Tensor Processing Units with code examples. A Tensor Processing Unit (TPU) is an application-specific integrated circuit (ASIC) designed to accelerate ML workloads. The TPUs available in TensorFlow are custom-developed from the ground up by the Google Brain team based on its plethora of experience and leadership in the ML community. Google Brain is a deep learning artificial intelligence (AI) research team at Google who research ways to make machines intelligent for the improvement of people’s lives.
- Published
- 2021
14. Scikit-Learn Classifier Tuning from Complex Training Sets
- Author
-
David Paper
- Subjects
Complex training ,Computer science ,business.industry ,Pattern recognition ,Artificial intelligence ,business ,Classifier (UML) - Abstract
Now that we have practiced tuning low-dimensional (or simple) data, we are ready to experiment tuning high-dimensional (or complex) data sets. Low-dimensional data consists of a limited number of features, whereas high-dimensional data consists of a very high number of features.
- Published
- 2019
15. Introduction to Scikit-Learn
- Author
-
David Paper
- Subjects
business.industry ,Computer science ,Unsupervised learning ,Artificial intelligence ,Python (programming language) ,Machine learning ,computer.software_genre ,business ,computer ,computer.programming_language - Abstract
We combine the Anaconda distribution with the popular Scikit-Learn library to demonstrate a wide range of supervised machine learning algorithms supplemented with unsupervised learning algorithms where appropriate. With clear examples, all written in Python, we demonstrate how these algorithms work to solve machine learning problems.
- Published
- 2019
16. Classification from Simple Training Sets
- Author
-
David Paper
- Subjects
Complex data type ,Data set ,Matrix (mathematics) ,Data element ,Simple (abstract algebra) ,business.industry ,Computer science ,Feature vector ,Pattern recognition ,Artificial intelligence ,Feature set ,business - Abstract
Classification from complex data is handled exactly as with simple data. Data is loaded into feature set X and target y. X data is composed of a matrix of vectors where each vector represents a data element and y data is composed of a vector of targets. However, complex data is composed of a high number of features (hundreds to thousands). Such a data set is commonly referred to as one with a high-dimensional feature space. Text data is also complex because each document must be converted into vectors of numerical values suitable for machine learning algorithms.
- Published
- 2019
17. On the computability of agent-based workflows
- Author
-
Prashant Palvia, David Paper, and Wai Yin Mok
- Subjects
Information Systems and Management ,Dependency (UML) ,Linear programming ,Computer science ,business.industry ,Computability ,Petri net ,Management Information Systems ,Workflow technology ,Workflow ,Arts and Humanities (miscellaneous) ,Developmental and Educational Psychology ,Artificial intelligence ,State diagram ,Software engineering ,business ,Workflow management system ,Information Systems - Abstract
Workflow research is commonly concerned with optimization, modeling, and dependency. In this research, we however address a more fundamental issue. By modeling humans and machines as agents and making use of a theoretical computer and statecharts, we prove that many workflow problems do not have computer-based solutions. We also demonstrate a sufficient condition under which computers are able to solve these problems. We end by discussing the relationships between our research and Petri Nets, the multi-agent framework in the literature, linear programming and workflow verification.
- Published
- 2006
18. A Short Grammatical Outline of Pashto
- Author
-
Herbert Penzl, D. A. Shafeev, and Herbert H. Paper
- Subjects
Linguistics and Language ,Grammar ,Computer science ,Arabic ,business.industry ,media_common.quotation_subject ,Phonology ,computer.software_genre ,Language and Linguistics ,language.human_language ,Linguistics ,Romanization ,language ,Pashto ,Artificial intelligence ,business ,computer ,Natural language processing ,media_common ,Language research - Published
- 1965
19. Left Ventricle Quantification Challenge: A Comprehensive Comparison and Evaluation of Segmentation and Regression for Mid-Ventricular Short-Axis Cardiac MR Data
- Author
-
Georgios Tziritas, Yeonggul Jang, Jin Ma, Fumin Guo, Quanzheng Li, Tiancong Hua, Xiang Li, Lihong Liu, Angélica Atehortúa, James R. Clough, Zhiqiang Hu, Eric Kerfoot, Vicente Grau, Enzo Ferrante, Matthew Ng, Guanyu Yang, Mireille Garreau, Alejandro Debus, Elias Grinias, Jiahui Li, Wufeng Xue, Shuo Li, Wenjun Yan, Ilkay Oksuz, Hao Xu, Shenzhen University, Beijing University of Posts and Telecommunications (BUPT), Peking University [Beijing], King‘s College London, Istanbul Technical University (ITÜ), University of Oxford [Oxford], University of Toronto, Massachusetts General Hospital [Boston], University of Crete [Heraklion] (UOC), Fudan University [Shanghai], Universidad Nacional de Colombia [Bogotà] (UNAL), Laboratoire Traitement du Signal et de l'Image (LTSI), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National de la Santé et de la Recherche Médicale (INSERM), Centre de Recherche en Information Biomédicale sino-français (CRIBS), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Southeast University [Jiangsu]-Institut National de la Santé et de la Recherche Médicale (INSERM), Yonsei University, Universidad Nacional del Litoral [Santa Fe] (UNL), Laboratory of Image Science and Technology [Nanjing] (LIST), Southeast University [Jiangsu]-School of Computer Science and Engineering, University of Western Ontario (UWO), The paper is partially supported by the Natural Science Foundation of China under Grants 61801296. The workof Eric Kerfoot was supported by an EPSRC programmeGrant (EP/P001009/1) and the Wellcome EPSRC Centre for Medical Engineering at the School of Biomedical Engineering and Imaging Sciences, Kings College London (WT203148/Z/16/Z). The work of Angelica Atehortua was supported by Colciencias-Colombia, Grant No. 647 (2015 call for National PhD studies) and Université de Rennes 1. The work of Alejandro Debus was supported by the Santa Fe Science, Technology and Innovation Agency (AS ACTEI), Government of the Province of Santa Fe, through Project AC-00010-18,Resolution N 117/14., University of Oxford, Université de Rennes (UR)-Institut National de la Santé et de la Recherche Médicale (INSERM), Université de Rennes (UR)-Southeast University [Jiangsu]-Institut National de la Santé et de la Recherche Médicale (INSERM), and Jonchère, Laurent
- Subjects
Short axis ,[SDV.IB.IMA]Life Sciences [q-bio]/Bioengineering/Imaging ,Computer science ,Heart Ventricles ,Magnetic Resonance Imaging, Cine ,Article ,030218 nuclear medicine & medical imaging ,03 medical and health sciences ,0302 clinical medicine ,[SDV.MHEP.CSC]Life Sciences [q-bio]/Human health and pathology/Cardiology and cardiovascular system ,Health Information Management ,medicine ,Humans ,Segmentation ,Electrical and Electronic Engineering ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing ,Ground truth ,Cardiac cycle ,business.industry ,Heart ,Pattern recognition ,Image segmentation ,Magnetic Resonance Imaging ,Regression ,[SDV.MHEP.CSC] Life Sciences [q-bio]/Human health and pathology/Cardiology and cardiovascular system ,Computer Science Applications ,[SDV.IB.IMA] Life Sciences [q-bio]/Bioengineering/Imaging ,medicine.anatomical_structure ,Ventricle ,030220 oncology & carcinogenesis ,Artificial intelligence ,business ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,Cardiac phase ,Biotechnology - Abstract
Automatic quantification of the left ventricle (LV) from cardiac magnetic resonance (CMR) images plays an important role in making the diagnosis procedure efficient, reliable, and alleviating the laborious reading work for physicians. Considerable efforts have been devoted to LV quantification using different strategies that include segmentation-based (SG) methods and the recent direct regression (DR) methods. Although both SG and DR methods have obtained great success for the task, a systematic platform to benchmark them remains absent because of differences in label information during model learning. In this paper, we conducted an unbiased evaluation and comparison of cardiac LV quantification methods that were submitted to the Left Ventricle Quantification (LVQuan) challenge, which was held in conjunction with the Statistical Atlases and Computational Modeling of the Heart (STACOM) workshop at the MICCAI 2018. The challenge was targeted at the quantification of 1) areas of LV cavity and myocardium, 2) dimensions of the LV cavity, 3) regional wall thicknesses (RWT), and 4) the cardiac phase, from mid-ventricle short-axis CMR images. First, we constructed a public quantification dataset Cardiac-DIG with ground truth labels for both the myocardium mask and these quantification targets across the entire cardiac cycle. Then, the key techniques employed by each submission were described. Next, quantitative validation of these submissions were conducted with the constructed dataset. The evaluation results revealed that both SG and DR methods can offer good LV quantification performance, even though DR methods do not require densely labeled masks for supervision. Among the 12 submissions, the DR method LDAMT offered the best performance, with a mean estimation error of 301 mm $^2$ for the two areas, 2.15 mm for the cavity dimensions, 2.03 mm for RWTs, and a 9.5% error rate for the cardiac phase classification. Three of the SG methods also delivered comparable performances. Finally, we discussed the advantages and disadvantages of SG and DR methods, as well as the unsolved problems in automatic cardiac quantification for clinical practice applications.
- Published
- 2021
20. Explaining Deep Learning Models for Speech Enhancement
- Author
-
Sunit Sivasankaran, Emmanuel Vincent, Dominique Fohr, Microsoft Corporation [Redmond], Microsoft Corporation [Redmond, Wash.], Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL), This work was made with the support of the French National Research Agency, in the framework of the project VOCADOM 'Robust voice command adapted to the user and to the context for AAL' (ANR-16-CE33-0006). Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as otherorganizations (see https://www.grid5000.fr)., ANR-16-CE33-0006,VOCADOM,Commande vocale robuste adaptée à la personne et au contexte pour l'autonomie à domicile(2016), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
Artificial neural network ,business.industry ,Computer science ,Deep learning ,Speech recognition ,Word error rate ,020206 networking & telecommunications ,Context (language use) ,02 engineering and technology ,explainable AI ,Speech enhancement ,030507 speech-language pathology & audiology ,03 medical and health sciences ,Noise ,feature attribution ,[INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] ,Robustness (computer science) ,[INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD] ,0202 electrical engineering, electronic engineering, information engineering ,Feature (machine learning) ,speech enhancement ,Artificial intelligence ,0305 other medical science ,business - Abstract
International audience; We consider the problem of explaining the robustness of neural networks used to compute time-frequency masks for speech enhancement to mismatched noise conditions. We employ the Deep SHapley Additive exPlanations (DeepSHAP) feature attribution method to quantify the contribution of every timefrequency bin in the input noisy speech signal to every timefrequency bin in the output time-frequency mask. We define an objective metric-referred to as the speech relevance scorethat summarizes the obtained SHAP values and show that it correlates with the enhancement performance, as measured by the word error rate on the CHiME-4 real evaluation dataset. We use the speech relevance score to explain the generalization ability of three speech enhancement models trained using synthetically generated speech-shaped noise, noise from a professional sound effects library, or real CHiME-4 noise. To the best of our knowledge, this is the first study on neural network explainability in the context of speech enhancement.
- Published
- 2021
21. Modeling and training strategies for language recognition systems
- Author
-
Raphaël Duroselle, Denis Jouvet, Md. Sahidullah, Irina Illina, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Experiments presented in this paper were carried out using the Grid5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr). This work has been partly funded by the French Direction Générale de l’Armement., Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)
- Subjects
business.industry ,Computer science ,multi-task training ,020206 networking & telecommunications ,02 engineering and technology ,computer.software_genre ,endto-end speech recognition ,Training (civil) ,language recognition ,030507 speech-language pathology & audiology ,03 medical and health sciences ,bottleneck features ,0202 electrical engineering, electronic engineering, information engineering ,[INFO]Computer Science [cs] ,Artificial intelligence ,0305 other medical science ,business ,computer ,Natural language processing ,Language recognition - Abstract
International audience; Automatic speech recognition is complementary to language recognition. The language recognition systems exploit this complementarity by using frame-level bottleneck features extracted from neural networks trained with a phone recognition task. Recent methods apply frame-level bottleneck features extracted from an end-to-end sequence-to-sequence speech recognition model. In this work, we study an integrated approach of the training of the speech recognition feature extractor and language recognition modules. We show that for both classical phone recognition and end-to-end sequence-to-sequence features, sequential training of the two modules is not the optimal strategy. The feature extractor can be improved by supervision with the language identification loss, either in a fine-tuning step or in a multi-task training framework. Besides, we notice that end-to-end sequence-to-sequence bottleneck features are on par with classical phone recognition bottleneck features without requiring a forced alignment of the signal with target tokens. However, for sequence-to-sequence, the architecture of the model seems to play an important role; the Conformer architectures leads to much better results than the conventional stacked DNNs approach; and can even be trained directly with the LID module in an end-to-end approach.
- Published
- 2021
22. HInT: Hybrid and Incremental Type Discovery for Large RDF Data Sources
- Author
-
Georgia Troullinou, Kenza Kellou-Menouer, Zoubida Kedad, Dimitris Plexousakis, Nikolaos Kardoulakis, Haridimos Kondylakis, Données et algorithmes pour une ville intelligente et durable - DAVID (DAVID), Université de Versailles Saint-Quentin-en-Yvelines (UVSQ), Hellenic Foundation for Research and Innovation, ΕΛ.ΙΔ.Ε.Κ: 1147, and Work reported in this paper has been partially supported by the Hellenic Foundation for Research and Innovation (H.F.R.I.) under the '2nd Call for H.F.R.I. Research Projects to support Post-Doctoral Researchers' (iQARuS Project No 1147)
- Subjects
Exploit ,LSH ,Process (engineering) ,Computer science ,media_common.quotation_subject ,02 engineering and technology ,Machine learning ,computer.software_genre ,RDF ,020204 information systems ,0202 electrical engineering, electronic engineering, information engineering ,Quality (business) ,[INFO]Computer Science [cs] ,media_common ,Complement (set theory) ,business.industry ,Hybrid type discovery ,Linked data ,computer.file_format ,Automatic summarization ,Schema (genetic algorithms) ,020201 artificial intelligence & image processing ,Incrementality ,Artificial intelligence ,business ,computer - Abstract
International audience; The rapid explosion of linked data has resulted into many weakly structured and incomplete data sources, where typing information might be missing. On the other hand, type information is essential for a number of tasks such as query answering, integration, summarization and partitioning. Existing approaches for type discovery, either completely ignore type declarations available in the dataset (implicit type discovery approaches), or rely only on existing types, in order to complement them (explicit type enrichment approaches). Implicit type discovery approaches are based on instance grouping, which requires an exhaustive comparison between the instances. This process is expensive and not incremental. Explicit type enrichment approaches on the other hand, are not able to identify new types and they can not process data sources that have little or no schema information. In this paper, we present HInT, the first incremental and hybrid type discovery system for RDF datasets, enabling type discovery in datasets where type declarations are missing. To achieve this goal, we incrementally identify the patterns of the various instances, we index and then group them to identify the types. During the processing of an instance, our approach exploits its type information, if available, to improve the quality of the discovered types by guiding the classification of the new instance in the correct group and by refining the groups already built. We analytically and experimentally show that our approach dominates in terms of efficiency, competitors from both worlds, implicit type discovery and explicit type enrichment while outperforming them in most of the cases in terms of quality.
- Published
- 2021
23. Mixture modeling for identifying subtypes in disease course mapping
- Author
-
Stanley Durrleman, Pierre-Emmanuel Poulet, Algorithms, models and methods for images and signals of the human brain (ARAMIS), Sorbonne Université (SU)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut du Cerveau et de la Moëlle Epinière = Brain and Spine Institute (ICM), Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Sorbonne Université-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Sorbonne Université-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), This research has received funding from the program 'Investissements d’avenir' ANR-10-IAIHU-06. This work was also funded in part by the French government under management of Agence Nationale de la Recherche as part of the 'Investissements d’avenir' program, reference ANR-19-P3IA-0001 (PRAIRIE 3IA Institute)., Aasa Feragen, Stefan Sommer, Julia Schnabel, Mads Nielsen, ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Institut du Cerveau et de la Moëlle Epinière = Brain and Spine Institute (ICM), Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut du Cerveau = Paris Brain Institute (ICM), Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Institut du Cerveau = Paris Brain Institute (ICM), Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Sorbonne Université (SU)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), This paper is funded in part by grant number 678304 (ERC), 826421 (TVB-Cloud) from H2020 programme, and ANR-10-IAIHU-06 (IHU ICM), ANR-19-P3IA-0001 (PRAIRIE) and ANR-19-JPW2-000 (E-DADS) from ANR., Poulet, Pierre-Emmanuel, PaRis Artificial Intelligence Research InstitutE - - PRAIRIE2019 - ANR-19-P3IA-0001 - P3IA - VALID, Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-CHU Pitié-Salpêtrière [AP-HP], Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
Computer science ,Population ,Stochastic approximation ,Machine learning ,computer.software_genre ,01 natural sciences ,Synthetic data ,010104 statistics & probability ,03 medical and health sciences ,0302 clinical medicine ,[STAT.ML]Statistics [stat]/Machine Learning [stat.ML] ,Alzheimer's disease subtypes ,mixed-effect models ,0101 mathematics ,Cognitive decline ,mixture models ,education ,Cluster analysis ,Non-linear mixed-effect model ,ComputingMilieux_MISCELLANEOUS ,Mixture model ,Ground truth ,education.field_of_study ,MCMC-SAEM ,business.industry ,Contrast (statistics) ,Disease course mapping ,Disease progression modelling ,[STAT.ML] Statistics [stat]/Machine Learning [stat.ML] ,Artificial intelligence ,business ,computer ,[STAT.ME]Statistics [stat]/Methodology [stat.ME] ,030217 neurology & neurosurgery - Abstract
International audience; Disease modeling techniques summarize the possible trajectories of progression from multimodal and longitudinal data. These techniques often assume that individuals form a homogeneous cluster, thus ignoring possible disease subtypes within the population. We extend a non-linear mixed-effect model used for disease course mapping with a mixture framework. We jointly estimate model parameters and subtypes with a tempered version of a stochastic approximation of the Expectation Maximisation algorithm. We show that our model recovers the ground truth parameters from synthetic data, in contrast to the naive solution consisting in post hoc clustering of individual parameters from a one-class model. Applications to Alzheimer's disease data allows the unsupervised identification of disease subtypes associated with distinct relationship between cognitive decline and progression of imaging and biological biomarkers.
- Published
- 2021
24. Single-molecule localization microscopy
- Author
-
Markus Sauer, Florian Schueder, Melike Lakadamyali, Gerti Beliu, Melina Theoni Gyparaki, Suliana Manley, Ralf Jungmann, Juliette Griffié, Mickaël Lelek, Christophe Zimmer, Imagerie et Modélisation - Imaging and Modeling, Institut Pasteur [Paris] (IP)-Centre National de la Recherche Scientifique (CNRS), University of Pennsylvania, University of Würzburg = Universität Würzburg, Ludwig Maximilian University [Munich] (LMU), Max-Planck-Institut für Biochemie = Max Planck Institute of Biochemistry (MPIB), Max-Planck-Gesellschaft, Ecole Polytechnique Fédérale de Lausanne (EPFL), Institut Pasteur [Paris] (IP)-Centre National de la Recherche Scientifique (CNRS)-Université Paris Cité (UPCité), C.Z. acknowledges funding by Institut Pasteur, Fondation pour la Recherche Médicale (grant DEQ 20150331762), Région Ile de France, Agence Nationale de la Recherche and Investissement d’Avenir grant ANR-16-CONV-0005. M.La. acknowledges funding from the National Institutes of Health/National Institutes of General Medical Sciences (NIH/NIGMS) under grant RO1 GM133842-01. G.B. and M.S. acknowledge funding by the German Research Foundation (DFG) (SA829/19-1) and the European Regional Development Fund (EFRE project ‘Center for Personalized Molecular Immunotherapy’). F.S. and R.J. acknowledge support by the DFG through SFB1032 (project A11) and the Max Planck Society. J.G. and S.M. acknowledge funding by the European Union’s H2020 programme under the Marie Skłodowska-Curie grant BALTIC (to J.G.) and ERC Piko (to S.M.)., The authors apologize to the authors of numerous papers that could not be cited owing to limited space. M.Le. and C.Z. thank B. Lelandais for excellent comments on the manuscript and M. Singh for acquiring the image shown in Fig. 3b., ANR-16-CONV-0005,INCEPTION,Institut Convergences pour l'étude de l'Emergence des Pathologies au Travers des Individus et des populatiONs(2016), Institut Pasteur [Paris]-Centre National de la Recherche Scientifique (CNRS), University of Pennsylvania [Philadelphia], Max Planck Institute of Biochemistry (MPIB), and Institut Pasteur [Paris]-Centre National de la Recherche Scientifique (CNRS)-Université Paris Cité (UPC)
- Subjects
Single molecule localization ,light-microscopy ,Computer science ,business.industry ,General Medicine ,Article ,General Biochemistry, Genetics and Molecular Biology ,optical reconstruction microscopy ,diffraction-limit ,Fluorescent labelling ,3-dimensional superresolution ,live-cell ,correlative superresolution fluorescence ,Microscopy ,Time course ,Image acquisition ,colocalization analysis ,[SDV.SPEE]Life Sciences [q-bio]/Santé publique et épidémiologie ,Computer vision ,Artificial intelligence ,business ,living cells ,intramolecular spirocyclization ,Image resolution ,electron-microscopy - Abstract
Single-molecule localization microscopy (SMLM) describes a family of powerful imaging techniques that dramatically improve spatial resolution over standard, diffraction-limited microscopy techniques and can image biological structures at the molecular scale. In SMLM, individual fluorescent molecules are computationally localized from diffraction-limited image sequences and the localizations are used to generate a super-resolution image or a time course of super-resolution images, or to define molecular trajectories. In this Primer, we introduce the basic principles of SMLM techniques before describing the main experimental considerations when performing SMLM, including fluorescent labelling, sample preparation, hardware requirements and image acquisition in fixed and live cells. We then explain how low-resolution image sequences are computationally processed to reconstruct super-resolution images and/or extract quantitative information, and highlight a selection of biological discoveries enabled by SMLM and closely related methods. We discuss some of the main limitations and potential artefacts of SMLM, as well as ways to alleviate them. Finally, we present an outlook on advanced techniques and promising new developments in the fast-evolving field of SMLM. We hope that this Primer will be a useful reference for both newcomers and practitioners of SMLM. This Primer explains the central concepts of single-molecule localization microscopy (SMLM) before discussing experimental considerations regarding fluorophores, optics and data acquisition, processing and analysis. The Primer further describes recent high-impact discoveries made by SMLM techniques and concludes by discussing emerging methodologies.
- Published
- 2021
25. Optimizing Motor Intention Detection With Deep Learning: Towards Management of Intraoperative Awareness
- Author
-
Laurent Bougrain, Oleksii Avilov, Anton Popov, Sébastien Rimbert, National Technical University of Ukraine 'Kyiv Polytechnic Institute' [Kiev], Analysis and modeling of neural systems by a system neuroscience approach (NEUROSYS), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Complex Systems, Artificial Intelligence & Robotics (LORIA - AIS), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL), Oleksii Avilov was supported by scholarship from the French Embassy to Ukraine while working on this topic at the NEUROSYS team at LORIA (Université de Lorraine/CNRS/Inria), Nancy, France. Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (https://www.grid5000.fr)., Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
0209 industrial biotechnology ,Computer science ,Biomedical Engineering ,motor imagery AAGA: accidental awareness during general anesthesia ,02 engineering and technology ,Intention ,Electroencephalography ,[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] ,Intraoperative Awareness ,020901 industrial engineering & automation ,Motor imagery ,median nerve stimulation ,Deep Learning ,[INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] ,intraoperative awareness during general anesthesia ,0202 electrical engineering, electronic engineering, information engineering ,medicine ,[INFO.INFO-IM]Computer Science [cs]/Medical Imaging ,Functional electrical stimulation ,Humans ,Brain-computer interface (BCI) ,Artificial neural network ,medicine.diagnostic_test ,Median nerve stimulation ,business.industry ,electroencephalogram (EEG) ,Deep learning ,[SCCO.NEUR]Cognitive science/Neuroscience ,Pattern recognition ,Linear discriminant analysis ,Median nerve ,medicine.anatomical_structure ,machine learning ,Frontal lobe ,Brain-Computer Interfaces ,Imagination ,020201 artificial intelligence & image processing ,Artificial intelligence ,business ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,Algorithms ,Motor cortex - Abstract
International audience; Objective: This article shows the interest in deep learning techniques to detect motor imagery (MI) from raw electroencephalographic (EEG) signals when a functional electrical stimulation is added or not. Impacts of electrode montages and bandwidth are also reported. The perspective of this work is to improve the detection of intraoperative awareness during general anesthesia. Methods: Various architectures of EEGNet were investigated to optimize MI detection. They have been compared to the state-of-the-art classifiers in Brain-Computer Interfaces (based on Riemannian geometry, linear discriminant analysis), and other deep learning architectures (deep convolution network, shallow convolutional network). EEG data were measured from 22 participants performing motor imagery with and without median nerve stimulation. Results: The proposed architecture of EEGNet reaches the best classification accuracy (83.2%) and false-positive rate (FPR 19.0%) for a setup with only six electrodes over the motor cortex and frontal lobe and for an extended 4-38 Hz EEG frequency range while the subject is being stimulated via a median nerve. Configurations with a larger number of electrodes result in higher accuracy (94.5%) and FPR (6.1%) for 128 electrodes (and respectively 88.0% and 12.9% for 13 electrodes).Conclusion: The present work demonstrates that using an extended EEG frequency band and a modified EEGNet deep neural network increases the accuracy of MI detection when used with as few as 6 electrodes which include frontal channels. Significance: The proposed method contributes to the development of Brain-Computer Interface systems based on MI detection from EEG.
- Published
- 2021
26. Does Infant-Directed Speech Help Phonetic Learning? A Machine Learning Investigation
- Author
-
Reiko Mazuka, Emmanuel Dupoux, Bogdan Ludusan, RIKEN Center for Brain Science [Wako] (RIKEN CBS), RIKEN - Institute of Physical and Chemical Research [Japon] (RIKEN), Universität Bielefeld = Bielefeld University, Department of Psychology and Neuroscience, Duke University [Durham], Apprentissage machine et développement cognitif (CoML), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de sciences cognitives et psycholinguistique (LSCP), CIFAR program in Learning in Machines & Brains CIFAR LMB program, The research reported in this paper was partly funded by JSPS Grant-in-Aid for Scientific Research (16H06319, 20H05617) and MEXT Grant-in-Aid on Innovative Areas #4903 (Co-creative Language Evolution), 17H06382 to R. Mazuka. The work of E. Dupoux in his EHESS role was supported by the European Research Council (ERC-2011-AdG-295810 BOOTPHON) the Agence Nationale pour la Recherche (ANR-10-LABX-0087 IEC, ANR-10-IDEX0001-02 PSL*, ANR-19-P3IA-0001 PRAIRIE 3IA Institute), and CIFAR (Learning in Machines and Brain). Part of the work was conducted while E. Dupoux was a visiting scientist at DeepMind and Facebook. B. Ludusan was also supported by the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement no. 799022., ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017), ANR-10-IDEX-0001,PSL,Paris Sciences et Lettres(2010), European Project: 295810,EC:FP7:ERC,ERC-2011-ADG_20110406,BOOTPHON(2012), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
- Subjects
Functional role ,Adult ,Computer science ,Cognitive Neuroscience ,Experimental and Cognitive Psychology ,Machine learning ,computer.software_genre ,Infant-directed speech ,050105 experimental psychology ,Read speech ,Machine Learning ,03 medical and health sciences ,0302 clinical medicine ,[INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] ,[STAT.ML]Statistics [stat]/Machine Learning [stat.ML] ,Artificial Intelligence ,Robustness (computer science) ,Phonetics ,Concept learning ,Vowel ,Humans ,Speech ,0501 psychology and cognitive sciences ,Speech variability ,Hyperarticulation ,business.industry ,05 social sciences ,Phonetic learning ,Infant ,Reading ,Speech Perception ,Artificial intelligence ,business ,computer ,Adult-directed speech ,030217 neurology & neurosurgery - Abstract
A prominent hypothesis holds that by speaking to infants in infant-directed speech (IDS) as opposed to adult-directed speech (ADS), parents help them learn phonetic categories. Specifically, two characteristics of IDS have been claimed to facilitate learning: hyperarticulation, which makes the categories more separable, and variability, which makes the generalization more robust. Here, we test the separability and robustness of vowel category learning on acoustic representations of speech uttered by Japanese adults in ADS, IDS (addressed to 18- to 24-month olds), or read speech (RS). Separability is determined by means of a distance measure computed between the five short vowel categories of Japanese, while robustness is assessed by testing the ability of six different machine learning algorithms trained to classify vowels to generalize on stimuli spoken by a novel speaker in ADS. Using two different speech representations, we find that hyperarticulated speech, in the case of RS, can yield better separability, and that increased between-speaker variability in ADS can yield, for some algorithms, more robust categories. However, these conclusions do not apply to IDS, which turned out to yield neither more separable nor more robust categories compared to ADS inputs. We discuss the usefulness of machine learning algorithms run on real data to test hypotheses about the functional role of IDS. © 2021 The Authors. Cognitive Science published by Wiley Periodicals LLC on behalf of Cognitive Science Society (CSS).
- Published
- 2020
27. Evolution Control for parallel ANN-assisted simulation-based optimization application to Tuberculosis Transmission Control
- Author
-
Mohand-Said Mezmaz, Nouredine Melab, Romain Ragonnet, Guillaume Briffoteaux, Daniel Tuyttens, Optimisation de grande taille et calcul large échelle (BONUS), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), University of Mons [Belgium] (UMONS), Monash University [Melbourne], Laboratoire d'Informatique Fondamentale de Lille (LIFL), Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS), and Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr).
- Subjects
Artificial Neural Network ,Computer Networks and Communications ,Computer science ,Monte Carlo method ,Context (language use) ,02 engineering and technology ,[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] ,Simulation-based optimization ,Surrogate-assisted Optimization ,0202 electrical engineering, electronic engineering, information engineering ,Massively parallel ,Dropout (neural networks) ,Artificial neural network ,Evolution Control ,business.industry ,Deep learning ,020206 networking & telecommunications ,[INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO] ,Supercomputer ,Computer engineering ,Hardware and Architecture ,020201 artificial intelligence & image processing ,Artificial intelligence ,Massively Parallel Computing ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,business ,Software ,Simulation - Abstract
International audience; In many optimal design searches, the function to optimise is a simulator that is computationally expensive. While current High Performance Computing (HPC) methods are not able to solve such problems efficiently, parallelism can be coupled with approximate models (surrogates or meta-models) that imitate the simulator in timely fashion to achieve better results. This combined approach reduces the number of simulations thanks to surrogate use whereas the remaining evaluations are handled by supercomputers. While the surrogates' ability to limit computational times is very attractive, integrating them into the over-arching optimization process can be challenging. Indeed, it is critical to address the major trade-off between the quality (precision) and the efficiency (execution time) of the resolution. In this article, we investigate Evolution Controls (ECs) which are strategies that define the alternation between the simulator and the surrogate within the optimization process. We propose a new EC based on the prediction uncertainty obtained from Monte Carlo Dropout (MCDropout), a technique originally dedicated to quantifying uncertainty in deep learning. Investigations of such uncertainty-aware ECs remain uncommon in surrogate-assisted evolutionary optimization. In addition, we use parallel computing in a complementary way to address the high computational burden. Our new strategy is implemented in the context of a pioneering application to Tuberculosis Transmission Control. The reported results show that the MCDropout-based EC coupled with massively parallel computing outperforms strategies previously proposed in the field of surrogate-assisted optimization.
- Published
- 2020
28. Unsupervised regularization of the embedding extractor for robust language identification
- Author
-
Raphaël Duroselle, Irina Illina, Denis Jouvet, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr). This work has been partly funded by the French Direction Générale de l'Armement., and Grid'5000
- Subjects
Language identification ,business.industry ,Computer science ,Pattern recognition ,[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] ,Regularization (mathematics) ,Extractor ,Maximum mean discrepancy ,Embedding ,Labeled data ,Artificial intelligence ,business ,Classifier (UML) ,Test data - Abstract
International audience; State-of-the-art spoken language identification systems are constituted of three modules: a frame-level feature extractor, a segment-level embedding extractor and a final classifier. The performance of these systems degrades when facing mismatch between training and testing data. Most domain adaptation methods focus on adaptation of the final classifier. In this article , we propose a model-based unsupervised domain adaptation of the segment-level embedding extractor. The approach consists in a modification of the loss function used for training the embedding extractor. We introduce a regularization term based on the maximum mean discrepancy loss. Experiments were performed on the RATS corpus with transmission channel mismatch between telephone and radio channels. We obtained the same language identification performance as supervised training on the target domains but without using labeled data from these domains.
- Published
- 2020
29. Interdisciplinary Research in Artificial Intelligence: Challenges and Opportunities
- Author
-
Remy Kusters, Dusan Misevic, Hugues Berry, Antoine Cully, Yann Le Cunff, Loic Dandoy, Natalia Díaz-Rodríguez, Marion Ficher, Jonathan Grizou, Alice Othmani, Themis Palpanas, Matthieu Komorowski, Patrick Loiseau, Clément Moulin Frier, Santino Nanini, Daniele Quercia, Michele Sebag, Françoise Soulié Fogelman, Sofiane Taleb, Liubov Tupikina, Vaibhav Sahu, Jill-Jênn Vie, Fatima Wehbi, Centre de Recherche Interdisciplinaire / Center for Research and Interdisciplinarity [Paris, France] (CRI), Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Paris Cité (UPCité), Artificial Evolution and Computational Biology (BEAGLE), Laboratoire d'InfoRmatique en Image et Systèmes d'information (LIRIS), Université Lumière - Lyon 2 (UL2)-École Centrale de Lyon (ECL), Université de Lyon-Université de Lyon-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Lumière - Lyon 2 (UL2)-École Centrale de Lyon (ECL), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de Biométrie et Biologie Evolutive - UMR 5558 (LBBE), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-VetAgro Sup - Institut national d'enseignement supérieur et de recherche en alimentation, santé animale, sciences agronomiques et de l'environnement (VAS)-Centre National de la Recherche Scientifique (CNRS)-VetAgro Sup - Institut national d'enseignement supérieur et de recherche en alimentation, santé animale, sciences agronomiques et de l'environnement (VAS)-Centre National de la Recherche Scientifique (CNRS), Imperial College London, Institut de Génétique et Développement de Rennes (IGDR), Université de Rennes (UR)-Centre National de la Recherche Scientifique (CNRS)-Structure Fédérative de Recherche en Biologie et Santé de Rennes ( Biosit : Biologie - Santé - Innovation Technologique ), Flowing Epigenetic Robots and Systems (Flowers), Inria Bordeaux - Sud-Ouest, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), École Nationale Supérieure de Techniques Avancées (ENSTA Paris), Laboratoire Images, Signaux et Systèmes Intelligents (LISSI), Université Paris-Est Créteil Val-de-Marne - Paris 12 (UPEC UP12), Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), Performance analysis and optimization of LARge Infrastructures and Systems (POLARIS), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'Informatique de Grenoble (LIG), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), Nokia Bell Labs [Cambridge], TAckling the Underspecified (TAU), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de Recherche en Informatique (LRI), CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Hub France IA, Nokia Bell Labs [Paris-Saclay], Scool (Scool), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Thanks to the Bettencourt Schueller Foundation long term partnership, the workshop that gave rise to this paper was partially supported by the funding from CRI Research Collaboratory, Institut National de la Santé et de la Recherche Médicale (INSERM)-Université de Paris (UP), Laboratoire de Biométrie et Biologie Evolutive - UMR 5558 (LBBE), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-VetAgro Sup - Institut national d'enseignement supérieur et de recherche en alimentation, santé animale, sciences agronomiques et de l'environnement (VAS)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-VetAgro Sup - Institut national d'enseignement supérieur et de recherche en alimentation, santé animale, sciences agronomiques et de l'environnement (VAS)-Centre National de la Recherche Scientifique (CNRS)-Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire d'InfoRmatique en Image et Systèmes d'information (LIRIS), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-École Centrale de Lyon (ECL), Université de Lyon-Université Lumière - Lyon 2 (UL2)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-École Centrale de Lyon (ECL), Université de Lyon-Université Lumière - Lyon 2 (UL2), Structure Fédérative de Recherche en Biologie et Santé de Rennes ( Biosit : Biologie - Santé - Innovation Technologique )-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Unité d'Informatique et d'Ingénierie des Systèmes (U2IS), École Nationale Supérieure de Techniques Avancées (ENSTA Paris)-École Nationale Supérieure de Techniques Avancées (ENSTA Paris), Université de Rennes 1 (UR1), and Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Centre National de la Recherche Scientifique (CNRS)-Structure Fédérative de Recherche en Biologie et Santé de Rennes ( Biosit : Biologie - Santé - Innovation Technologique )
- Subjects
Big Data ,0301 basic medicine ,[SCCO.COMP]Cognitive science/Computer science ,[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,03 medical and health sciences ,0302 clinical medicine ,Political science ,Health care ,Computer Science (miscellaneous) ,auditability ,education ,lcsh:T58.5-58.64 ,Ai education ,business.industry ,lcsh:Information technology ,interdisciplinary science ,Precision medicine ,artificial intelligence ,Transparency (behavior) ,ethics ,Variety (cybernetics) ,030104 developmental biology ,[INFO.INFO-MA]Computer Science [cs]/Multiagent Systems [cs.MA] ,Perspective ,Artificial intelligence ,business ,interpretability ,030217 neurology & neurosurgery ,Information Systems - Abstract
The use of artificial intelligence (AI) in a variety of research fields is speeding up multiple digital revolutions, from shifting paradigms in healthcare, precision medicine and wearable sensing, to public services and education offered to the masses around the world, to future cities made optimally efficient by autonomous driving. When a revolution happens, the consequences are not obvious straight away, and to date, there is no uniformly adapted framework to guide AI research to ensure a sustainable societal transition. To answer this need, here we analyze three key challenges to interdisciplinary AI research, and deliver three broad conclusions: 1) future development of AI should not only impact other scientific domains but should also take inspiration and benefit from other fields of science, 2) AI research must be accompanied by decision explainability, dataset bias transparency as well as development of evaluation methodologies and creation of regulatory agencies to ensure responsibility, and 3) AI education should receive more attention, efforts and innovation from the educational and scientific communities. Our analysis is of interest not only to AI practitioners but also to other researchers and the general public as it offers ways to guide the emerging collaborations and interactions toward the most fruitful outcomes.
- Published
- 2020
30. Machines and Masterpieces: Predicting Prices in the Art Auction Market
- Author
-
Roman Kräussl, Gustavo Manso, Mathieu Aubry, Christophe Spaenjers, Laboratoire d'Informatique Gaspard-Monge (LIGM), Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM), Department of Economics, Tilburg University [Netherlands], HEC Research Paper Series, and Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS)
- Subjects
asset valuation ,History ,Polymers and Plastics ,Computer science ,Big data ,Machine learning ,computer.software_genre ,Industrial and Manufacturing Engineering ,big data ,0502 economics and business ,Common value auction ,Price level ,auctions ,050207 economics ,Business and International Management ,JEL: C - Mathematical and Quantitative Methods/C.C5 - Econometric Modeling/C.C5.C50 - General ,Valuation (finance) ,JEL: D - Microeconomics/D.D4 - Market Structure, Pricing, and Design/D.D4.D44 - Auctions ,050208 finance ,Artificial neural network ,Ex-ante ,business.industry ,05 social sciences ,experts ,Hedonic pricing ,TheoryofComputation_GENERAL ,JEL: Z - Other Special Topics/Z.Z1 - Cultural Economics • Economic Sociology • Economic Anthropology/Z.Z1.Z11 - Economics of the Arts and Literature ,machine learning ,[SHS.GESTION]Humanities and Social Sciences/Business administration ,Artificial intelligence ,business ,computer ,JEL: G - Financial Economics/G.G1 - General Financial Markets/G.G1.G12 - Asset Pricing • Trading Volume • Bond Interest Rates - Abstract
We study the accuracy and usefulness of automated (i.e., machine-generated) valuations for illiquid and heterogeneous real assets. We assemble a database of 1.1 million paintings auctioned between 2008 and 2015. We use a popular machine-learning technique - neural networks - to develop a pricing algorithm based on both non-visual and visual artwork characteristics. Our out-of-sample valuations predict auction prices dramatically better than valuations based on a standard hedonic pricing model. Moreover, they help explaining price levels and sale probabilities even after conditioning on auctioneers' pre-sale estimates. Machine learning is particularly helpful for assets that are associated with high price uncertainty. It can also correct human experts' systematic biases in expectations formation - and identify ex ante situations in which such biases are likely to arise.
- Published
- 2020
31. Parallel Surrogate-assisted Optimization: Batched Bayesian Neural Network-assisted GA versus q-EGO
- Author
-
Guillaume Briffoteaux, Romain Ragonnet, Jan Gmys, Nouredine Melab, Mohand Mezmaz, Maxime Gobert, Daniel Tuyttens, Optimisation de grande taille et calcul large échelle (BONUS), Inria Lille - Nord Europe, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), University of Mons [Belgium] (UMONS), Monash University [Melbourne], Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Institut de Mathématiques [Mons], Université de Mons (UMons), Laboratoire d'Informatique Fondamentale de Lille (LIFL), Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS), and Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientific interest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr).
- Subjects
General Computer Science ,Computer science ,General Mathematics ,Evolutionary algorithm ,02 engineering and technology ,[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] ,Machine learning ,computer.software_genre ,Surrogate model ,Genetic algorithm ,Surrogate-assisted Optimization ,0202 electrical engineering, electronic engineering, information engineering ,Massively parallel ,Global optimization ,Artificial neural network ,Bayesian Optimization ,business.industry ,05 social sciences ,Bayesian optimization ,Evolutionary Algorithm ,050301 education ,[INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO] ,Efficient Global Optimization ,Benchmark (computing) ,020201 artificial intelligence & image processing ,Artificial intelligence ,Massively Parallel Computing ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,business ,0503 education ,computer ,Simulation - Abstract
International audience; Surrogate-based optimization is widely used to deal with long-running black-box simulation-based objective functions. Actually, the use of a surrogate model such as Kriging or Artificial Neural Network allows to reduce the number of calls to the CPU time-intensive simulator. Bayesian optimization uses the ability of surrogates to provide useful information to help guiding effectively the optimization process. In this paper, the Efficient Global Optimization (EGO) reference framework is challenged by a Bayesian Neural Network-assisted Genetic Algorithm, namely BNN-GA. The Bayesian Neural Network (BNN) surrogate is chosen for its ability to provide an uncertainty measure of the prediction that allows to compute the Expected Improvement of a candidate solution in order to improve the exploration of the objective space. BNN is also more reliable than Kriging models for high-dimensional problems and faster to set up thanks to its incremental training. In addition, we propose a batch-based approach for the parallelization of BNN-GA that is challenged by a parallel version of EGO, called q-EGO. Parallel computing is a highly important complementary way (to surrogates) to deal with the computational burden of simulation-based optimization. The comparison of the two parallel approaches is experimentally performed through several benchmark functions and two real-world problems within the scope of Tuberculosis Transmission Control (TBTC). The study presented in this paper proves that parallel batched BNN-GA is a viable alternative to q-EGO approaches being more suitable for high-dimensional problems, parallelization impact, bigger databases and moderate search budgets. Moreover, a significant improvement of the solutions is obtained for the two TBTC problems tackled.
- Published
- 2020
32. Foreground-Background Ambient Sound Scene Separation
- Author
-
Michel Olvera, Romain Serizel, Emmanuel Vincent, Gilles Gasso, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes (LITIS), Université Le Havre Normandie (ULH), Normandie Université (NU)-Normandie Université (NU)-Université de Rouen Normandie (UNIROUEN), Normandie Université (NU)-Institut national des sciences appliquées Rouen Normandie (INSA Rouen Normandie), Institut National des Sciences Appliquées (INSA)-Normandie Université (NU)-Institut National des Sciences Appliquées (INSA), This work was made with the support of the French National Research Agency, in the framework of the project LEAUDS 'Learning to understandaudio scenes' (ANR-18-CE23-0020). Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientificinterest group hosted by Inria and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr)., GRID5000, ANR-18-CE23-0020,LEAUDS,Apprentissage statistique pour la compréhension de scènes audio(2018), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL), Institut national des sciences appliquées Rouen Normandie (INSA Rouen Normandie), Normandie Université (NU)-Institut National des Sciences Appliquées (INSA)-Normandie Université (NU)-Institut National des Sciences Appliquées (INSA)-Université de Rouen Normandie (UNIROUEN), Normandie Université (NU)-Université Le Havre Normandie (ULH), Normandie Université (NU), and ANR-18-CE23-0020,LEAUDS,LEARNING TO UNDERSTAND AUDIO SCENES(2018)
- Subjects
Normalization (statistics) ,Signal Processing (eess.SP) ,FOS: Computer and information sciences ,Sound (cs.SD) ,Computer Science - Machine Learning ,Computer science ,Generalization ,Ambient noise level ,02 engineering and technology ,Computer Science - Sound ,Machine Learning (cs.LG) ,Signal-to-noise ratio ,[INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,Audio and Speech Processing (eess.AS) ,0202 electrical engineering, electronic engineering, information engineering ,ambient sound scenes ,FOS: Electrical engineering, electronic engineering, information engineering ,Foreground-background ,Computer vision ,generalization ability ,Electrical Engineering and Systems Science - Signal Processing ,Sound (geography) ,Signal processing ,geography ,geography.geographical_feature_category ,business.industry ,Deep learning ,deep learning ,audio source separation ,020206 networking & telecommunications ,Feature (computer vision) ,[INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD] ,020201 artificial intelligence & image processing ,Artificial intelligence ,business ,Electrical Engineering and Systems Science - Audio and Speech Processing - Abstract
International audience; Ambient sound scenes typically comprise multiple short events occurring on top of a somewhat stationary background. We consider the task of separating these events from the background, which we call foreground-background ambient sound scene separation. We propose a deep learning-based separation framework with a suitable feature normaliza-tion scheme and an optional auxiliary network capturing the background statistics, and we investigate its ability to handle the great variety of sound classes encountered in ambient sound scenes, which have often not been seen in training. To do so, we create single-channel foreground-background mixtures using isolated sounds from the DESED and Audioset datasets, and we conduct extensive experiments with mixtures of seen or unseen sound classes at various signal-to-noise ratios. Our experimental findings demonstrate the generalization ability of the proposed approach.
- Published
- 2020
33. The trainer, the verifier, the imitator: Three ways in which human platform workers support artificial intelligence
- Author
-
Paola Tubaro, Antonio A. Casilli, Marion Coville, Centre National de la Recherche Scientifique (CNRS), TAckling the Underspecified (TAU), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de Recherche en Informatique (LRI), CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Laboratoire de Recherche en Informatique (LRI), CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), Institutions et Dynamiques Historiques de l'Économie et de la Société (IDHES), Université Paris 1 Panthéon-Sorbonne (UP1)-Université Paris 8 Vincennes-Saint-Denis (UP8)-Université Paris Nanterre (UPN)-Université d'Évry-Val-d'Essonne (UEVE)-Centre National de la Recherche Scientifique (CNRS)-Ecole Normale Supérieure Paris-Saclay (ENS Paris Saclay), Sociologie Information-Communication Design (SID), Institut interdisciplinaire de l’innovation de Telecom Paris (I3 SES), Télécom ParisTech-Institut interdisciplinaire de l’innovation (I3), Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Télécom ParisTech-Institut interdisciplinaire de l’innovation (I3), Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS), Département Sciences Economiques et Sociales (SES), Télécom ParisTech, Institut Polytechnique de Paris (IP Paris), Institut d'Administration des Entreprises (IAE) - Poitiers (IAE Poitiers), Université de Poitiers, CEntre de REcherche en GEstion - EA 1722 (CEREGE), Université de Poitiers-Université de Poitiers-Université de Poitiers-La Rochelle Université (ULR), This paper presents results of a larger study called Digital Platform Labour (DiPLab), co-funded by Maison des Sciences de l’Homme Paris-Saclay, Force Ouvrière, a workers’ union, as part of a grant from Institut de recherches économiques et sociales (IRES), and France Stratégie, a service of the French Prime Minister’s office. The platform Foule Factory offered logistical support, and Inria provided complementary funding. For more information: http://diplab.eu, Ecole Normale Supérieure Paris-Saclay (ENS Paris Saclay)-Université d'Évry-Val-d'Essonne (UEVE)-Centre National de la Recherche Scientifique (CNRS)-Université Paris Nanterre (UPN)-Université Paris 8 Vincennes-Saint-Denis (UP8)-Université Paris 1 Panthéon-Sorbonne (UP1), Institut interdisciplinaire de l’innovation (I3, une unité mixte de recherche CNRS (UMR 9217)), Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Télécom ParisTech-MINES ParisTech - École nationale supérieure des mines de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Télécom ParisTech-MINES ParisTech - École nationale supérieure des mines de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Université de Poitiers-Université de Poitiers-Université de Poitiers-Université de La Rochelle (ULR), École polytechnique (X)-Télécom ParisTech-Mines Paris - PSL (École nationale supérieure des mines de Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-École polytechnique (X)-Télécom ParisTech-Mines Paris - PSL (École nationale supérieure des mines de Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), and École polytechnique (X)-Télécom ParisTech-MINES ParisTech - École nationale supérieure des mines de Paris
- Subjects
Information Systems and Management ,Computer science ,Trainer ,Big data ,050801 communication & media studies ,lcsh:A ,Library and Information Sciences ,micro-work ,Digital platform labour ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,0508 media and communications ,datafied production processes ,0502 economics and business ,050207 economics ,Focus (computing) ,[SHS.SOCIO]Humanities and Social Sciences/Sociology ,business.industry ,Communication ,05 social sciences ,artificial intelligence ,[SHS.ECO]Humanities and Social Sciences/Economics and Finance ,Computer Science Applications ,machine learning ,Artificial intelligence ,lcsh:General Works ,business ,Information Systems - Abstract
International audience; This paper sheds light on the role of digital platform labour in the development of today's artificial intelligence, predicated on data-intensive machine learning algorithms. Focus is on the specific ways in which outsourcing of data tasks to myriad 'micro-workers', recruited and managed through specialized platforms, powers virtual assistants, self-driving vehicles and connected objects. Using qualitative data from multiple sources, we show that micro-work performs a variety of functions, between three poles that we label, respectively, 'artificial intelligence preparation', 'artificial intelligence verification' and 'artificial intelligence impersonation'. Because of the wide scope of application of micro-work, it is a structural component of contemporary artificial intelligence production processes - not an ephemeral form of support that may vanish once the technology reaches maturity stage. Through the lens of micro-work, we prefigure the policy implications of a future in which data technologies do not replace human workforce but imply its marginalization and precariousness.
- Published
- 2020
34. Geometry description and mesh construction from medical imaging
- Author
-
Silvia Bertoluzza, Minh Phan, Michela Spagnuolo, Philippe Ricka, Micol Pennacchio, Giuseppe Patanè, Michele Giuliano Carlino, Institut de Mathématiques de Bordeaux (IMB), Université Bordeaux Segalen - Bordeaux 2-Université Sciences et Technologies - Bordeaux 1-Université de Bordeaux (UB)-Institut Polytechnique de Bordeaux (Bordeaux INP)-Centre National de la Recherche Scientifique (CNRS), Modeling Enablers for Multi-PHysics and InteractionS (MEMPHIS), Université Bordeaux Segalen - Bordeaux 2-Université Sciences et Technologies - Bordeaux 1-Université de Bordeaux (UB)-Institut Polytechnique de Bordeaux (Bordeaux INP)-Centre National de la Recherche Scientifique (CNRS)-Université Bordeaux Segalen - Bordeaux 2-Université Sciences et Technologies - Bordeaux 1-Université de Bordeaux (UB)-Institut Polytechnique de Bordeaux (Bordeaux INP)-Centre National de la Recherche Scientifique (CNRS)-Inria Bordeaux - Sud-Ouest, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Institut de Recherche Mathématique Avancée (IRMA), Centre National de la Recherche Scientifique (CNRS)-Université de Strasbourg (UNISTRA), Institut de Mathématiques de Toulon - EA 2134 (IMATH), Université de Toulon (UTLN), Istituto di Matematica Applicata e Tecnologie Informatiche (IMATI), Consiglio Nazionale delle Ricerche (CNR), Istituto di Matematica Applicata e Tecnologie Informatiche (IMATI-CNR), Consiglio Nazionale delle Ricerche [Roma] (CNR), his paper has been realized in the framework of ERC Project CHANGE, which has received funding from the EuropeanResearch Council (ERC) under the European Unions Horizon 2020 research and innovation programme (grant agreement No694515., European Project: 8509436(1986), Université Bordeaux Segalen - Bordeaux 2-Université Sciences et Technologies - Bordeaux 1 (UB)-Université de Bordeaux (UB)-Institut Polytechnique de Bordeaux (Bordeaux INP)-Centre National de la Recherche Scientifique (CNRS), Université Bordeaux Segalen - Bordeaux 2-Université Sciences et Technologies - Bordeaux 1 (UB)-Université de Bordeaux (UB)-Institut Polytechnique de Bordeaux (Bordeaux INP)-Centre National de la Recherche Scientifique (CNRS)-Université Bordeaux Segalen - Bordeaux 2-Université Sciences et Technologies - Bordeaux 1 (UB)-Université de Bordeaux (UB)-Institut Polytechnique de Bordeaux (Bordeaux INP)-Centre National de la Recherche Scientifique (CNRS)-Inria Bordeaux - Sud-Ouest, Université de Strasbourg (UNISTRA)-Centre National de la Recherche Scientifique (CNRS), and National Research Council of Italy | Consiglio Nazionale delle Ricerche (CNR)
- Subjects
T57-57.97 ,Applied mathematics. Quantitative methods ,Computer simulation ,Computer science ,business.industry ,Atlas (topology) ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Image segmentation ,Modular design ,Reference image ,N/A ,QA1-939 ,Medical imaging ,[INFO.INFO-IM]Computer Science [cs]/Medical Imaging ,A priori and a posteriori ,Computer vision ,[INFO]Computer Science [cs] ,Artificial intelligence ,business ,Mathematics ,ComputingMethodologies_COMPUTERGRAPHICS - Abstract
International audience; We present a new method for defining and meshing patient-specific domains from medical images. Our approach is based on an atlas image segmentation technique, and relies on the modular registration algorithm of S. Bertoluzza et al. [25]. The mesh of the patient-specific domain is generated by deforming the corresponding mesh on an a priori segmented and meshed reference image (the atlas). Our method aims at automating the process at the interface of medical imaging and numerical simulation, thus reducing the computational cost in those situations where simulations have to be managed on numerous medical images of similar patients.; Nous présentons une nouvelle méthode de reconnaissance et de maillage d’un domaine d’intérêt d’une image médicale. Notre approche se base sur une méthode de segmentation à partir d’un atlas, et dépend de la boîte à outils modulaire pour la co-registration d’images développée par S. Bertoluzza et al. [25]. Le maillage du domaine spécifique au patient est généré par déformation du maillage correspondant sur une image de référence segmentée a priori (l’atlas). Notre méthode vise à automatiser le processus à l’interface entre l’imagerie médicale et la simulation numérique, avec pour but de réduire le coût de calcul dans les situations dans lesquelles des simulations doivent être faites sur de nombreuses images similaires.
- Published
- 2020
35. Computing AES related-key differential characteristics with constraint programming
- Author
-
Christine Solnon, Marine Minier, David Gerault, Pascal Lafourcade, Université Clermont Auvergne [2017-2020] (UCA [2017-2020]), Laboratoire d'Informatique, de Modélisation et d'optimisation des Systèmes (LIMOS), SIGMA Clermont (SIGMA Clermont)-Université d'Auvergne - Clermont-Ferrand I (UdA)-Ecole Nationale Supérieure des Mines de St Etienne-Centre National de la Recherche Scientifique (CNRS)-Université Blaise Pascal - Clermont-Ferrand 2 (UBP), Cryptology, arithmetic : algebraic methods for better algorithms (CARAMBA), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Algorithms, Computation, Image and Geometry (LORIA - ALGO), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Geometry Processing and Constrained Optimization (M2DisCo), Laboratoire d'InfoRmatique en Image et Systèmes d'information (LIRIS), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-École Centrale de Lyon (ECL), Université de Lyon-Université Lumière - Lyon 2 (UL2)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Université Lumière - Lyon 2 (UL2), This research was conducted with the support of the FEDER program of 2014-2020, the region council of Auvergne-Rhône-Alpes, the GDR-IA, and the ANR (DeCrypt ANR-18-CE39-0007). We thank the reviewers for their comments that helped us improving the paper, Jérémie Detrey for implementing the C code that checks the completeness of the sets xorEq_l, Charles Prud’homme and Jean-Guillaume Fages for their technical support on the use of Choco, and Neng-Fa Zhou for his technical support on the use of Picat., ANR-16-IDEX-0001,CAP 20-25,CAP 20-25(2016), ANR-18-CE39-0007,DeCrypt,Langage Déclaratif pour la cryptographie symétrique(2018), Université Blaise Pascal - Clermont-Ferrand 2 (UBP)-Université d'Auvergne - Clermont-Ferrand I (UdA)-SIGMA Clermont (SIGMA Clermont)-Ecole Nationale Supérieure des Mines de St Etienne (ENSM ST-ETIENNE)-Centre National de la Recherche Scientifique (CNRS), Université Lumière - Lyon 2 (UL2)-École Centrale de Lyon (ECL), Université de Lyon-Université de Lyon-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Lumière - Lyon 2 (UL2)-École Centrale de Lyon (ECL), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
Linguistics and Language ,Differential cryptanalysis ,Theoretical computer science ,Optimization problem ,Computer science ,02 engineering and technology ,Encryption ,Language and Linguistics ,law.invention ,Constraint Programming ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,[INFO.INFO-CR]Computer Science [cs]/Cryptography and Security [cs.CR] ,Artificial Intelligence ,law ,020204 information systems ,0202 electrical engineering, electronic engineering, information engineering ,Constraint programming ,Block cipher ,AES ,business.industry ,Advanced Encryption Standard ,Key (cryptography) ,020201 artificial intelligence & image processing ,Cryptanalysis ,business - Abstract
International audience; Cryptanalysis aims at testing the properties of encryption processes, and this usually implies solving hard optimization problems. In this paper, we focus on related-key differential attacks for the Advanced Encryption Standard (AES), which is the encryption standard for block ciphers. To mount these attacks, cryptanalysts need to solve the optimal related-key differential characteristic problem. Dedicated approaches do not scale well for this problem, and need weeks to solve its hardest instances. In this paper, we improve existing Constraint Programming (CP) approaches for computing optimal related-key differential characteristics: we add new constraints that detect inconsistencies sooner, and we introduce a new decomposition of the problem in two steps. These improvements allow us to compute all optimal related-key differential characteristics for AES-128, AES-192 and AES-256 in a few hours.
- Published
- 2020
36. Automatic Waypoint Generation to Improve Robot Navigation Through Narrow Spaces
- Author
-
Javier Gonzalez-Jimenez, Jose-Raul Ruiz-Sarmiento, Cipriano Galindo, Javier Monroy, Francisco-Angel Moreno, Machine Perception and Intelligent Robotics Group (MAPIR), [Moreno,FA, Monroy,J, Ruiz-Sarmiento,JR, Galindo,C, Gonzalez-Jimenez,J] Machine Perception and Intelligent Robotics Group (MAPIR), Dept. of System Engineering and Automation Biomedical Research Institute of Malaga (IBIMA), University of Malaga, Málaga, Spain., and This work has been supported by the research projects WISER (DPI2017-84827-R), funded by the Spanish Government and the European Regional Development’s Funds (FEDER), MoveCare (ICT-26-2016b-GA-732158), funded by the European H2020 program, and by a postdoc contract from the I-PPIT program of the University of Malaga. The publication of this paper has been funded by the University of Malaga.
- Subjects
0209 industrial biotechnology ,Traverse ,Phenomena and Processes::Mathematical Concepts::Algorithms [Medical Subject Headings] ,Computer science ,Chemicals and Drugs::Inorganic Chemicals::Free Radicals::Reactive Oxygen Species [Medical Subject Headings] ,waypoint generation ,Real-time computing ,Information Science::Information Science::Computing Methodologies::Artificial Intelligence::Robotics [Medical Subject Headings] ,02 engineering and technology ,lcsh:Chemical technology ,Biochemistry ,Article ,Analytical Chemistry ,Task (project management) ,Waypoint generation ,Organisms::Eukaryota::Animals::Chordata::Vertebrates::Mammals::Primates::Haplorhini::Catarrhini::Hominidae::Humans [Medical Subject Headings] ,Waypoint ,robot deployment ,020901 industrial engineering & automation ,Information Science::Information Science::Computing Methodologies::Artificial Intelligence [Medical Subject Headings] ,Navigation assistant ,robot navigation ,Artificial Intelligence ,mobile robots ,robot localization ,0202 electrical engineering, electronic engineering, information engineering ,Mobile robots ,Mapa ,lcsh:TP1-1185 ,Electrical and Electronic Engineering ,Instrumentation ,business.industry ,Robot localization ,Robotics ,Mobile robot ,Atomic and Molecular Physics, and Optics ,Inteligencia artificial ,Trajectory ,Robot navigation ,Map ,Robot ,020201 artificial intelligence & image processing ,Noise (video) ,Artificial intelligence ,business ,Robótica ,Robot deployment - Abstract
In domestic robotics, passing through narrow areas becomes critical for safe and effective robot navigation. Due to factors like sensor noise or miscalibration, even if the free space is sufficient for the robot to pass through, it may not see enough clearance to navigate, hence limiting its operational space. An approach to facing this is to insert waypoints strategically placed within the problematic areas in the map, which are considered by the robot planner when generating a trajectory and help to successfully traverse them. This is typically carried out by a human operator either by relying on their experience or by trial-and-error. In this paper, we present an automatic procedure to perform this task that: (i) detects problematic areas in the map and (ii) generates a set of auxiliary navigation waypoints from which more suitable trajectories can be generated by the robot planner. Our proposal, fully compatible with the robotic operating system (ROS), has been successfully applied to robots deployed in different houses within the H2020 MoveCare project. Moreover, we have performed extensive simulations with four state-of-the-art robots operating within real maps. The results reveal significant improvements in the number of successful navigations for the evaluated scenarios, demonstrating its efficacy in realistic situations.
- Published
- 2019
- Full Text
- View/download PDF
37. Bidirectional LSTM Autoencoder for Sequence Based Anomaly Detection in Cyber Security
- Author
-
Ashima Chawla, Brian Lee, Paul Jacob, Sheila Fallon, and This project reported in this paper has received funding from the European Union Horizon 2020 research and innovation programme under grant agreement No. 700071 for the PROTECTIVE project.
- Subjects
business.industry ,Computer science ,Pattern recognition ,Autoencoders ,Autoencoder ,Embeddings ,System call ,Modeling and Simulation ,Anomaly detection ,Artificial intelligence ,Host based intrusion ,business ,Software ,CuDNNLSTM ,Software Research Institute AIT ,Sequence (medicine) - Abstract
Cyber-security is concerned with protecting information, a vital asset in today’s world. The volume of data that is generated can be usefully analyzed when cyber-security systems are effectively implemented with the aid of software support. Our approach is to determine normal behavior of a system based on sequences of system call traces made by the kernel processes in the system. This paper describes a robust and computationally efficient anomaly based host based intrusion detection system using an Encoder-Decoder mechanism. Using CuDNNLSTM networks, it is possible to obtain a set of comparable results with reduced training times. The Bidirectional Encoder and a unidirectional Decoder is trained on normal call sequences in the ADFA-LD dataset. Intrusion Detection is evaluated based on determining the probability of a sequence being reconstructed by the model yes
- Published
- 2019
38. Towards Portable Online Prediction of Network Utilization using MPI-level Monitoring
- Author
-
Emmanuel Jeannot, Franck Cappello, Shu Mei Tseng, Bogdan Nicolae, Aparna Chandramowlishwaran, George Bosilca, University of California [Irvine] (UCI), University of California, Argonne National Laboratory [Lemont] (ANL), The University of Tennessee [Knoxville], Topology-Aware System-Scale Data Management for High-Performance Computing (TADAAM), Laboratoire Bordelais de Recherche en Informatique (LaBRI), Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Inria Bordeaux - Sud-Ouest, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), This research was supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. Department of Energy Office of Science and the National Nuclear Security Administration. This material was based upon work supported by the U.S. Department of Energy, Office of Science, under contract DE-AC02-06CH11357,and by the National Science Foundation under Grant No. #1664142. The experimentspresented in this paper were carried out using the Grid’5000/ALADDIN-G5K experimental testbed, an initiative of the French Ministry of Research through the ACI GRID incentive action, INRIA, CNRS and RENATER and other contributing partners (see http://www.grid5000.fr/)., University of California [Irvine] (UC Irvine), University of California (UC), and Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS)-Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS)-Inria Bordeaux - Sud-Ouest
- Subjects
Artificial neural network ,business.industry ,Computer science ,Network monitoring ,Deep learning ,Distributed computing ,010103 numerical & computational mathematics ,010501 environmental sciences ,Prediction of resource utilization ,01 natural sciences ,Timeseries forecasting ,Work stealing ,Online learning ,Leverage (statistics) ,Artificial intelligence ,0101 mathematics ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,business ,0105 earth and related environmental sciences - Abstract
International audience; Stealing network bandwidth helps a variety of HPC runtimes and services to run additional operations in the background without negatively affecting the applications. A key ingredient to make this possible is an accurate prediction of the future network utilization, enabling the runtime to plan the background operations in advance, such as to avoid competing with the application for network bandwidth. In this paper, we propose a portable deep learning predictor that only uses the information available through MPI introspection to construct a recurrent sequence-to-sequence neural network capable of forecasting network utilization. We leverage the fact that most HPC applications exhibit periodic behaviors to enable predictions far into the future (at least the length of a period). Our on-line approach does not have an initial training phase, it continuously improves itself during application execution without incurring significant computational overhead. Experimental results show better accuracy and lower computational overhead compared with the state-of-the-art on two representative applications.
- Published
- 2019
39. Classification of Broadcast News Audio Data Employing Binary Decision Architecture
- Author
-
Peter Feciľak, Anton Čižmár, Jozef Juhar, Jozef Vavrek, and The research in this paper was supported by the Ministry of Education, Science, Research and Sport of the Slovak Republic under the project VEGA 2/0197/15 and the Slovak Research and Development Agency under the project APVV-15-0517.
- Subjects
0209 industrial biotechnology ,Computational complexity theory ,Computer Networks and Communications ,Computer science ,Binary number ,68T10 ,02 engineering and technology ,computer.software_genre ,Multiclass classification ,020901 industrial engineering & automation ,0202 electrical engineering, electronic engineering, information engineering ,Architecture ,Support vector machine, audio classification, broadcast news data, binary decision trees, binary decision architecture ,other areas of Computing and Informatics ,Binary decision diagram ,business.industry ,020208 electrical & electronic engineering ,Pattern recognition ,Support vector machine ,ComputingMethodologies_PATTERNRECOGNITION ,Computational Theory and Mathematics ,Hardware and Architecture ,Data mining ,Artificial intelligence ,Decision table ,business ,computer ,Classifier (UML) ,Software - Abstract
A novel binary decision architecture (BDA) for broadcast news audio classification task is presented in this paper. The idea of developing such architecture came from the fact that the appropriate combination of multiple binary classifiers for two-class discrimination problem can reduce a miss-classification error without rapid increase in computational complexity. The core element of classification architecture is represented by a binary decision (BD) algorithm that performs discrimination between each pair of acoustic classes, utilizing two types of decision functions. The first one is represented by a simple rule-based approach in which the final decision is made according to the value of selected discrimination parameter. The main advantage of this solution is relatively low processing time needed for classification of all acoustic classes. The cost for that is low classification accuracy. The second one employs support vector machine (SVM) classifier. In this case, the overall classification accuracy is conditioned by finding the optimal parameters for decision function resulting in higher computational complexity and better classification performance. The final form of proposed BDA is created by combining four BD discriminators supplemented by decision table. The effectiveness of proposed BDA, utilizing rule-based approach and the SVM classifier, is compared with two most popular strategies for multiclass classification, namely the binary decision trees (BDT) and the One-Against-One SVM (OAOSVM). Experimental results show that the proposed classification architecture can decrease the overall classification error in comparison with the BDT architecture. On the contrary, an optimization technique for selecting the optimal set of training data is needed in order to overcome the OAOSVM.
- Published
- 2017
40. Real-time analysis of the behaviour of groups of mice via a depth-sensing camera and machine learning
- Author
-
Philippe Faure, Jean-Christophe Olivo-Marin, Fabrice de Chaumont, Nicolas Torquet, Albane Imbert, Stephane Dallongeville, Anne-Marie Le Sourd, Thomas Bourgeron, Elodie Ey, Thierry Legou, Thibault Lagache, Analyse d'images biologiques - Biological Image Analysis (BIA), Institut Pasteur [Paris]-Centre National de la Recherche Scientifique (CNRS), Génétique humaine et fonctions cognitives - Human Genetics and Cognitive Functions (GHFC (UMR_3571 / U-Pasteur_1)), Institut Pasteur [Paris]-Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS), Neurosciences Paris Seine (NPS), Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut de Biologie Paris Seine (IBPS), Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche et Innovation Technologique (CITECH), Institut Pasteur [Paris], Laboratoire Parole et Langage (LPL), Aix Marseille Université (AMU)-Centre National de la Recherche Scientifique (CNRS), This work was partially funded by the Institut Pasteur, the Bettencourt-Schueller Foundation, the Cognacq–Jay Foundation, the Conny–Maeva Foundation, the ERANET–NEURON SYNPATHY program, the Agence Nationale de la Recherche through grant number ANR-10-LABX-62-IBEID, France-BioImaging infrastructure through grant number ANR-10-INBS-04 and the INCEPTION program through grant number ANR-16-CONV-0005, the Centre National de la Recherche Scientifique, the University Paris Diderot, the BioPsy Labex, the Institut National du Cancer through grant number TABAC-16–022, the Foundation for Medical Research (Equipe DEQ20130326488), the Innovative Medicines Initiative Joint Undertaking through grant agreement number 115300, resources of which are composed of financial contributions from the European Union’s Seventh Framework Program (FP7/2007–2013) and EFPIA companies in kind contribution., The authors thank Y. Archambeau and P. Ollivon at the workshop of the Institut Pasteur for building the first 12 setups and advising on hardware, W. Meiniel for the mathematical proof for decisions of head/tail probability, Microsoft France for their technical support, P. Spinicelli for optical engineering and reading of the paper, R. Marée for machine learning support, B. König for advice and reading of biological experiments, J. N. Crawley for reading and providing comments on the manuscript, A. Barmpoutis for providing us with the early Kinect 2 driver and support, N. Chenouard for driving the use of the machine learning solution, P. Dugast for drawing the mice in the different behavioural events, A. Engelberg for checking the English, S. Wagner and R. Accardi for RFID advice, M. Marim for website development, and X. Montagutelli and M. Bérard for animal facility support., Analyse d'images biologiques - BioImage Analysis (AIQ), Centre National de la Recherche Scientifique (CNRS)-Université Paris Diderot - Paris 7 (UPD7)-Institut Pasteur [Paris], Neuroscience Paris Seine (NPS), Centre National de la Recherche Scientifique (CNRS)-Institut de Biologie Paris Seine (IBPS), Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Pierre et Marie Curie - Paris 6 (UPMC), Centre National de la Recherche Scientifique (CNRS)-Aix Marseille Université (AMU), Institut Pasteur [Paris] (IP)-Centre National de la Recherche Scientifique (CNRS), Institut Pasteur [Paris] (IP)-Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS), Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut de Biologie Paris Seine (IBPS), Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS), Institut Pasteur [Paris] (IP), ANR-16-CONV-0005,INCEPTION,Institut Convergences pour l'étude de l'Emergence des Pathologies au Travers des Individus et des populatiONs(2016), Centre National de la Recherche Scientifique (CNRS)-Institut Pasteur [Paris], Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS)-Institut Pasteur [Paris], and Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
0301 basic medicine ,Male ,[SDV.NEU.NB]Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC]/Neurobiology ,Biomedical Engineering ,Video Recording ,Medicine (miscellaneous) ,Bioengineering ,Nerve Tissue Proteins ,Mouse tracking ,Biology ,Machine learning ,computer.software_genre ,Machine Learning ,03 medical and health sciences ,Mice ,0302 clinical medicine ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,Animals ,Autistic Disorder ,Real time analysis ,Social Behavior ,Mice, Knockout ,Behavior, Animal ,business.industry ,Microfilament Proteins ,Computer Science Applications ,Disease Models, Animal ,030104 developmental biology ,Phenotype ,Mutation ,Identification (biology) ,Female ,Artificial intelligence ,business ,computer ,030217 neurology & neurosurgery ,Biotechnology ,Behavioral Research - Abstract
Preclinical studies of psychiatric disorders use animal models to investigate the impact of environmental factors or genetic mutations on complex traits such as decision-making and social interactions. Here, we introduce a method for the real-time analysis of the behaviour of mice housed in groups of up to four over several days and in enriched environments. The method combines computer vision through a depth-sensing infrared camera, machine learning for animal and posture identification, and radio-frequency identification to monitor the quality of mouse tracking. It tracks multiple mice accurately, extracts a list of behavioural traits of both individuals and the groups of mice, and provides a phenotypic profile for each animal. We used the method to study the impact of Shank2 and Shank3 gene mutations—mutations that are associated with autism—on mouse behaviour. Characterization and integration of data from the behavioural profiles of Shank2 and Shank3 mutant female mice revealed their distinctive activity levels and involvement in complex social interactions. A method that combines a depth-sensing camera and machine learning can track the movements of up to four mice in real time and for several days, extracting both individual and group behavioural traits.
- Published
- 2019
41. Automatic Discovery of Families of Network Generative Processes
- Author
-
Telmo Menezes, Camille Roth, Centre Marc Bloch (CMB), Ministère de l'Europe et des Affaires étrangères (MEAE)-Bundesministerium für Bildung und Forschung-Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.)-Centre National de la Recherche Scientifique (CNRS), This paper has been partially supported by the 'Algodiv'' grant (ANR-15-CE38-0001) funded by the ANR (French National Agency of Research)., ANR-12-CORD-0018,Algopol,Politique des algorithmes(2012), and ANR-15-CE38-0001,ALGODIV,Algodiv: Recommandation algorithmique et diversité des informations du web(2015)
- Subjects
Computer science ,Network science ,Genetic Programming ,Complex networks ,Genetic programming ,[INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE] ,01 natural sciences ,[INFO.INFO-SI]Computer Science [cs]/Social and Information Networks [cs.SI] ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,03 medical and health sciences ,[INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] ,0103 physical sciences ,Selection (linguistics) ,Evolutionary computations ,010306 general physics ,Artificial Intelligence/Machine Learning ,030304 developmental biology ,Network model ,0303 health sciences ,[SHS.STAT]Humanities and Social Sciences/Methods and statistics ,[SHS.SOCIO]Humanities and Social Sciences/Sociology ,business.industry ,Complex network ,Machine Learning ML ,Network formation ,Computational social sciences ,Artificial intelligence ,Social network Analysis SNA ,business ,Symbolic regression ,Generative grammar - Abstract
International audience; Designing plausible network models typically requires scholars to form a priori intuitions on the key drivers of network formation. Oftentimes, these intuitions are supported by the statistical estimation of a selection of network evolution processes which will form the basis of the model to be developed. Machine learning techniques have lately been introduced to assist the automatic discovery of generative models. These approaches may more broadly be described as "symbolic regression", where fundamental network dynamic functions, rather than just parameters, are evolved through genetic programming. This chapter first aims at reviewing the principles, efforts and the emerging literature in this direction, which is very much aligned with the idea of creating artificial scientists. Our contribution then aims more specifically at building upon an approach recently developed by us [Menezes & Roth, 2014] in order to demonstrate the existence of families of networks that may be described by similar generative processes. In other words, symbolic regression may be used to group networks according to their inferred genotype (in terms of generative processes) rather than their observed phenotype (in terms of statistical/topological features). Our empirical case is based on an original data set of 238 anonymized ego-centered networks of Facebook friends, further yielding insights on the formation of sociability networks.
- Published
- 2019
42. Linear Search by a Pair of Distinct-Speed Robots
- Author
-
Tomasz Kociumaka, Jurek Czyzowicz, David Ilcinkas, Leszek Gąsieniec, Ralf Klasing, Dominik Pająk, Evangelos Bampas, School of of Electrical and Computer Engineering [Athens] (School of E.C.E), National Technical University of Athens [Athens] (NTUA), Département d'Informatique et d'Ingénierie (DII), Université du Québec en Outaouais (UQO), Department of Computer Science [Liverpool], University of Liverpool, Laboratoire Bordelais de Recherche en Informatique (LaBRI), Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB), University of Warsaw (UW), Institute of informatics [Wrocław], See paper for details., ANR-11-BS02-0014,DISPLEXITY,Calculabilité et complexité en distribué(2011), ANR-13-JS02-0002,MACARON,Bouger et Calculer: Agents, Robots et Réseaux(2013), ANR-16-CE40-0023,DESCARTES,Abstraction modulaire pour le calcul distribué(2016), ANR-10-IDEX-0003,IDEX BORDEAUX,Initiative d'excellence de l'Université de Bordeaux(2010), Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory, Pajak, Dominik S, Laboratoire d'informatique Fondamentale de Marseille (LIF), Aix Marseille Université (AMU)-École Centrale de Marseille (ECM)-Centre National de la Recherche Scientifique (CNRS), Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS), Institute of Informatics [Warsaw], Faculty of Mathematics, Informatics, and Mechanics [Warsaw] (MIMUW), University of Warsaw (UW)-University of Warsaw (UW), and Centre National de la Recherche Scientifique (CNRS)-École Centrale de Marseille (ECM)-Aix Marseille Université (AMU)
- Subjects
Engineering ,General Computer Science ,Computer science ,different speeds ,[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS] ,0211 other engineering and technologies ,0102 computer and information sciences ,02 engineering and technology ,Bang-bang robot ,linear search ,01 natural sciences ,Article ,Computer Science::Robotics ,Position (vector) ,mobile robots ,0202 electrical engineering, electronic engineering, information engineering ,Cartesian coordinate robot ,Point (geometry) ,Computer vision ,[INFO]Computer Science [cs] ,Online algorithm ,Simulation ,Linear search ,021103 operations research ,business.industry ,Applied Mathematics ,Mobile robot ,Computer Science Applications ,010201 computation theory & mathematics ,group search ,Moment (physics) ,Line (geometry) ,Robot ,020201 artificial intelligence & image processing ,Artificial intelligence ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,business - Abstract
Two mobile robots are initially placed at the same point on an infinite line. Each robot may move on the line in either direction not exceeding its maximal speed. The robots need to find a stationary target placed at an unknown location on the line. The search is completed when both robots arrive at the target point. The target is discovered at the moment when either robot arrives at its position. The robot knowing the placement of the target may communicate it to the other robot. We look for the algorithm with the shortest possible search time (i.e. the worst-case time at which both robots meet at the target) measured as a function of the target distance from the origin (i.e. the time required to travel directly from the starting point to the target at unit velocity). We consider two standard models of communication between the robots, namely wireless communication and communication by meeting. In the case of communication by meeting, a robot learns about the target while sharing the same location with a robot possessing this knowledge. We propose here an optimal search strategy for two robots including the respective lower bound argument, for the full spectrum of their maximal speeds. This extends the main result of Chrobak et al. (in: Italiano, Margaria-Steffen, Pokorný, Quisquater, Wattenhofer (eds) Current trends in theory and practice of computer science, SOFSEM, 2015) referring to the exact complexity of the problem for the case when the speed of the slower robot is at least one third of the faster one. In the wireless communication model, a message sent by one robot is instantly received by the other robot, regardless of their current positions on the line. For this model, we design a strategy which is optimal whenever the faster robot is at most \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sqrt{17}+4\approx 8.123$$\end{document}17+4≈8.123 times faster than the slower one. We also prove that otherwise the wireless communication offers no advantage over communication by meeting.
- Published
- 2019
43. Enhancing Perceived Safety in Human–Robot Collaborative Construction Using Immersive Virtual Environments
- Author
-
Sangseok You, Lionel P. Robert, Jeonghwan Kim, SangHyun Lee, Vineet R. Kamat, Okinawa Institute of Science and Technology Graduate University (OIST), Institute for Computational Engineering and Sciences [Austin] (ICES), University of Texas at Austin [Austin], and HEC Research Paper Series
- Subjects
0209 industrial biotechnology ,Identification ,Computer science ,0211 other engineering and technologies ,Theoretical models ,02 engineering and technology ,Robot Acceptance Safety Model (RASM) ,Virtual reality ,Trust ,Human–robot interaction ,020901 industrial engineering & automation ,Human–computer interaction ,021105 building & construction ,0502 economics and business ,Set (psychology) ,Masonry ,GeneralLiterature_REFERENCE(e.g.,dictionaries,encyclopedias,glossaries) ,Construction automation ,Civil and Structural Engineering ,Perceived safety ,Immersive Virtual ,business.industry ,05 social sciences ,Robotics ,Building and Construction ,Environment (IVE) ,Human–Robot Work Collaboration (HRWC) ,Identification (information) ,Work (electrical) ,Control and Systems Engineering ,Robot ,[SHS.GESTION]Humanities and Social Sciences/Business administration ,Intention to Work with Robot ,Artificial intelligence ,Safety ,Immersive virtual environment ,business ,Humanoid robot ,050203 business & management ,Virtual prototyping ,Team - Abstract
Advances in robotics now permit humans to work collaboratively with robots. However, humans often feel unsafe working alongside robots. Our knowledge of how to help humans overcome this issue is limited by two challenges. One, it is difficult, expensive and time-consuming to prototype robots and set up various work situations needed to conduct studies in this area. Two, we lack strong theoretical models to predict and explain perceived safety and its influence on human–robot work collaboration (HRWC). To address these issues, we introduce the Robot Acceptance Safety Model (RASM) and employ immersive virtual environments (IVEs) to examine perceived safety of working on tasks alongside a robot. Results from a between-subjects experiment done in an IVE show that separation of work areas between robots and humans increases perceived safety by promoting team identification and trust in the robot. In addition, the more participants felt it was safe to work with the robot, the more willing they were to work alongside the robot in the future.
- Published
- 2018
44. Films based on crosslinked TEMPO-oxidized cellulose and predictive analysis via machine learning
- Author
-
Jouni Paltakari, Merve Özkan, Alp Karakoç, Orlando J. Rojas, Maryam Borghei, Paper Converting and Packaging, Department of Bioproducts and Biosystems, Aalto-yliopisto, and Aalto University
- Subjects
Materials science ,Oxidized cellulose ,lcsh:Medicine ,02 engineering and technology ,010402 general chemistry ,Machine learning ,computer.software_genre ,01 natural sciences ,Polyvinyl alcohol ,Article ,chemistry.chemical_compound ,Transmittance ,Surface roughness ,Cellulose ,lcsh:Science ,Multidisciplinary ,business.industry ,lcsh:R ,021001 nanoscience & nanotechnology ,Flexible electronics ,0104 chemical sciences ,chemistry ,Nanofiber ,lcsh:Q ,Wetting ,Artificial intelligence ,0210 nano-technology ,business ,computer - Abstract
We systematically investigated the effect of film-forming polyvinyl alcohol and crosslinkers, glyoxal and ammonium zirconium carbonate, on the optical and surface properties of films produced from TEMPO-oxidized cellulose nanofibers (TOCNFs). In this regard, UV-light transmittance, surface roughness and wetting behavior of the films were assessed. Optimization was carried out as a function of film composition following the “random forest” machine learning algorithm for regression analysis. As a result, the design of tailor-made TOCNF-based films can be achieved with reduced experimental expenditure. We envision this approach to be useful in facilitating adoption of TOCNF for the design of emerging flexible electronics, and related platforms.
- Published
- 2018
45. Bootstrapping Q-Learning for Robotics from Neuro-Evolution Results
- Author
-
Matthieu Zimmer, Stéphane Doncieux, Architectures et modèles d'Adptation et de la cognition (AMAC), Institut des Systèmes Intelligents et de Robotique (ISIR), Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), This work has been supported by the FET project DREAM, that has received funding from the European Unions Horizon 2020 research and innovation programme under grant agreement No 640891. The authors would like to thank Olivier Sigaud for his comments on a first draft of the article. The data has been numerically analysed with the free software package GNU Octave and scikit-learn. Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientific interest group hosted by INRIA and including CNRS, RENATER and several Universities as well as other organizations (see https://www.grid5000.fr)., and GRID5000
- Subjects
Computer Science::Machine Learning ,0209 industrial biotechnology ,Learning classifier system ,business.industry ,Computer science ,Evolutionary robotics ,Q-learning ,Online machine learning ,Multi-task learning ,generation of representation during development ,02 engineering and technology ,transfer learning ,Robot learning ,robots with development and learning skills ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,020901 industrial engineering & automation ,Artificial Intelligence ,0202 electrical engineering, electronic engineering, information engineering ,Reinforcement learning ,020201 artificial intelligence & image processing ,Instance-based learning ,Artificial intelligence ,business ,Software - Abstract
International audience; Reinforcement learning problems are hard to solve in a robotics context as classical algorithms rely on discrete representations of actions and states, but in robotics both are continuous. A discrete set of actions and states can be defined, but it requires an expertise that may not be available, in particular in open environments. It is proposed to define a process to make a robot build its own representation for a reinforcement learning algorithm. The principle is to first use a direct policy search in the sensori-motor space, i.e. with no predefined discrete sets of states nor actions, and then extract from the corresponding learning traces discrete actions and identify the relevant dimensions of the state to estimate the value function. Once this is done, the robot can apply reinforcement learning (1) to be more robust to new domains and, if required, (2) to learn faster than a direct policy search. This approach allows to take the best of both worlds: first learning in a continuous space to avoid the need of a specific representation, but at a price of a long learning process and a poor generalization, and then learning with an adapted representation to be faster and more robust.
- Published
- 2017
46. Multi-dimensional signal approximation with sparse structured priors using split Bregman iterations
- Author
-
Michèle Sebag, Cédric Gouy-Pailler, Yoann Isaac, Quentin Barthélemy, Jamal Atif, Laboratoire d'analyse des données et d'intelligence des systèmes (LADIS), Département Métrologie Instrumentation & Information (DM2I), Laboratoire d'Intégration des Systèmes et des Technologies (LIST), Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Laboratoire d'Intégration des Systèmes et des Technologies (LIST), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay, TAckling the Underspecified (TAU), Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Mensia Technologies [Rennes], Mensia Technologies [Paris], Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), The work presented in this paper has been partially funded by DIGITEO under the Grant 2011-053D, Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay-Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA)), CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, and CentraleSupélec-Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
FOS: Computer and information sciences ,Signal processing ,Optimization problem ,Noise reduction ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,02 engineering and technology ,Machine Learning (cs.LG) ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,Recovery ,Computer Science - Data Structures and Algorithms ,Prior probability ,Regularization ,0202 electrical engineering, electronic engineering, information engineering ,Data Structures and Algorithms (cs.DS) ,Electrical and Electronic Engineering ,Selection ,Image restoration ,Mathematics ,business.industry ,Linear Inverse Problems ,020206 networking & telecommunications ,Pattern recognition ,Sparse approximation ,[INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA] ,Regression ,Minimization ,Computer Science - Learning ,Image-Restoration ,Control and Systems Engineering ,Statistical analysis ,020201 artificial intelligence & image processing ,[SDV.NEU]Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC] ,Computer Vision and Pattern Recognition ,Decomposition method (constraint satisfaction) ,Artificial intelligence ,Minification ,Lasso ,business ,Software ,Algorithms ,[MATH.MATH-NA]Mathematics [math]/Numerical Analysis [math.NA] - Abstract
This paper addresses the structurally constrained sparse decomposition of multi-dimensional signals onto overcomplete families of vectors, called dictionaries. The contribution of the paper is threefold. Firstly, a generic spatio-temporal regularization term is designed and used together with the standard ź 1 regularization term to enforce a sparse decomposition preserving the spatio-temporal structure of the signal. Secondly, an optimization algorithm based on the split Bregman approach is proposed to handle the associated optimization problem, and its convergence is analyzed. Our well-founded approach yields same accuracy as the other algorithms at the state of the art, with significant gains in terms of convergence speed. Thirdly, the empirical validation of the approach on artificial and real-world problems demonstrates the generality and effectiveness of the method. On artificial problems, the proposed regularization subsumes the Total Variation minimization and recovers the expected decomposition. On the real-world problem of electro-encephalography brainwave decomposition, the approach outperforms similar approaches in terms of P300 evoked potentials detection, using structured spatial priors to guide the decomposition. HighlightsA sparse structured decomposition method is proposed for multi-dimensional signals.Knowledge priors are encoded in a regularization to obtain plausible representations.The proposed split-Bregman based method outperforms counterparts in terms of speed.The approach is applied to EEG denoising for the extraction of P300 potentials.
- Published
- 2017
47. Blind estimation of blur in hyperspectral images
- Author
-
Benoit Vozel, Kacem Chehdi, Vladimir V. Lukin, Mykhail Uss, Mo Zhang, Sergey K. Abramov, Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Kharkov National University, This work has been co-funded by the Brittany Region and the Côtes d'Armor Department. The authors would like to thank all the authors of the semi-blind methods comparatively assessed in this paper for making their code publicly available., Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)
- Subjects
Point spread function ,Computer science ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Sparse distribution ,02 engineering and technology ,Regularization (mathematics) ,Augmented lagrangian ,[INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI] ,Signal-to-noise ratio ,0202 electrical engineering, electronic engineering, information engineering ,Alternating direction method of multipliers (admm) ,Regularization parameter ,Image restoration ,Total variation ,business.industry ,Cumulative distribution function ,Hyperspectral imaging ,020206 networking & telecommunications ,Pattern recognition ,Thresholding ,[SPI.TRON]Engineering Sciences [physics]/Electronics ,Kernel estimation ,020201 artificial intelligence & image processing ,Noise (video) ,Artificial intelligence ,business ,Salient edges ,Blind image restoration ,Remote sensing images - Abstract
International audience; Hyperspectral images acquired by remote sensing systems are generally degraded by noise and can be sometimes more severely degraded by blur. When no knowledge is available about the degradations present on the original image, blind restoration methods can only be considered. By blind, we mean absolutely no knowledge neither of the blur point spread function (PSF) nor the original latent channel and the noise level. In this study, we address the blind restoration of the degraded channels component-wise, according to a sequential scheme. For each degraded channel, the sequential scheme estimates the blur point spread function (PSF) in a first stage and deconvolves the degraded channel in a second and final stage by means of using the PSF previously estimated. We propose a new component-wise blind method for estimating effectively and accurately the blur point spread function. This method follows recent approaches suggesting the detection, selection and use of sufficiently salient edges in the current processed channel for supporting the regularized blur PSF estimation. Several modifications are beneficially introduced in our work. A new selection of salient edges through thresholding adequately the cumulative distribution of their corresponding gradient magnitudes is introduced. Besides, quasi-Automatic and spatially adaptive tuning of the involved regularization parameters is considered. To prove applicability and higher efficiency of the proposed method, we compare it against the method it originates from and four representative edge-sparsifying regularized methods of the literature already assessed in a previous work. Our attention is mainly paid to the objective analysis (via l1- norm) of the blur PSF error estimation accuracy. The tests are performed on a synthetic hyperspectral image. This synthetic hyperspectral image has been built from various samples from classified areas of a real-life hyperspectral image, in order to benefit from realistic spatial distribution of reference spectral signatures to recover after synthetic degradation. The synthetic hyperspectral image has been successively degraded with eight real blurs taken from the literature, each of a different support size. Conclusions, practical recommendations and perspectives are drawn from the results experimentally obtained. © 2017 SPIE.
- Published
- 2017
48. Subjective Contingencies and Limited Bayesian Updating
- Author
-
Stefania Minardi, Andrei Savochkin, Haldemann, Antoine, Groupement de Recherche et d'Etudes en Gestion à HEC (GREGH), Ecole des Hautes Etudes Commerciales (HEC Paris)-Centre National de la Recherche Scientifique (CNRS), Collegio Carlo Alberto, Università degli studi di Torino = University of Turin (UNITO), and HEC Research Paper Series
- Subjects
Economics and Econometrics ,Computer science ,media_common.quotation_subject ,Bayesian probability ,Bayesian inference ,Separable space ,Revealed preference ,0502 economics and business ,State space ,Relevance (information retrieval) ,050207 economics ,Representation (mathematics) ,Set (psychology) ,050205 econometrics ,media_common ,business.industry ,Welfare economics ,05 social sciences ,Representation (systemics) ,JEL: D - Microeconomics/D.D8 - Information, Knowledge, and Uncertainty/D.D8.D81 - Criteria for Decision-Making under Risk and Uncertainty ,understanding of uncertainty ,subjective state space ,non-Bayesian updating ,Confirmation bias ,[SHS.GESTION]Humanities and Social Sciences/Business administration ,Artificial intelligence ,business ,[SHS.GESTION] Humanities and Social Sciences/Business administration - Abstract
We depart from Savage's (1954) common state space assumption and introduce a model that allows for a subjective understanding of uncertainty. Within the revealed preference paradigm, we take the analyst's perspective and uniquely identify the agent's subjective state space via her preferences conditional on incoming information. According to our representation, the agent's subjective contingencies correspond to sets of the analyst's states and, as such, are coarse. The agent uses an additively separable utility with respect to her set of contingencies; and she adopts an updating rule that follows the Bayesian spirit but is limited by her perception of uncertainty. We illustrate the relevance of our theory with applications to the confirmatory bias and correlation neglect, as well as to optimal contract design.
- Published
- 2017
49. Classification Models Via Tabu Search: An Application to Early Stage Venture Classification
- Author
-
Canan Akdemir, Thomas B. Astebro, Samir Elhedhli, Department of Management Sciences, University of Waterloo [Waterloo], Joseph L. Rotman School of Management, University of Toronto, HEC Research Paper Series, Haldemann, Antoine, Groupement de Recherche et d'Etudes en Gestion à HEC (GREGH), and Ecole des Hautes Etudes Commerciales (HEC Paris)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
JEL: C - Mathematical and Quantitative Methods/C.C5 - Econometric Modeling/C.C5.C53 - Forecasting and Prediction Methods • Simulation Methods ,Mixed integer program ,Computer science ,media_common.quotation_subject ,Benders' decomposition ,Machine learning ,computer.software_genre ,decision heuristic ,JEL: C - Mathematical and Quantitative Methods/C.C4 - Econometric and Statistical Methods: Special Topics/C.C4.C45 - Neural Networks and Related Topics ,JEL: C - Mathematical and Quantitative Methods/C.C6 - Mathematical Methods • Programming Models • Mathematical and Simulation Modeling/C.C6.C63 - Computational Techniques • Simulation Modeling ,Artificial Intelligence ,Classification models ,tabu search ,Quality (business) ,Decision-making ,Selection (genetic algorithm) ,media_common ,Mathematics ,business.industry ,General Engineering ,early stage venture forecast ,large-scale mixed integer program ,Tabu search ,Computer Science Applications ,Data set ,classification ,[SHS.GESTION.STRAT]Humanities and Social Sciences/Business administration/domain_shs.gestion.strat ,[SHS.GESTION]Humanities and Social Sciences/Business administration ,Stage (hydrology) ,Data mining ,Artificial intelligence ,[SHS.GESTION] Humanities and Social Sciences/Business administration ,business ,computer ,Integer (computer science) - Abstract
We model the decision making process used by Experts at the Canadian Innovation Centre to classify early stage venture proposals based on potential commercial success. The decision is based on thirty-seven attributes that take values in { - 1 , 0 , 1 } . We adopt a conjunctive decision framework due to Astebro and Elhedhli (2005) that selects a subset of attributes and determines two threshold values: one for the maximum allowed negatives (n) and one for minimum required positives (p). A proposal is classified as a success if the number of positives is greater than or equal to p and the number of negatives is less than or equal to n over the selected attributes. Based on a data set of 561 observations, the selection of attributes and the determination of the threshold values is modeled as a large-scale mixed integer program. Two solution approaches are explored: Benders decomposition and Tabu search. The first, was very slow to converge, while the second provided high quality solutions quickly. Tabu search provides excellent classification accuracy for predicting commercial successes as well as replicating Experts’ forecasts, opening the venue for the use of Tabu search in scoring and classification problems.
- Published
- 2015
50. Disconnected Components Detection and Rooted Shortest-Path Tree Maintenance in Networks
- Author
-
Christian Glacet, Nicolas Hanusse, Colette Johnen, David Ilcinkas, Laboratoire Bordelais de Recherche en Informatique (LaBRI), Université de Bordeaux (UB)-Centre National de la Recherche Scientifique (CNRS)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB), ANR-16-CE40-0023,DESCARTES,Abstraction modulaire pour le calcul distribué(2016), ANR-16-CE25-0009,ESTATE,Auto-stabilisation et amélioration de la sûreté dans les environnements distribués évoluant dans le temps(2016), ANR-10-IDEX-0003,IDEX BORDEAUX,Initiative d'excellence de l'Université de Bordeaux(2010), See paper for details., and ANR-11-BS02-0014,DISPLEXITY,Calculabilité et complexité en distribué(2011)
- Subjects
Routing protocol ,Leader election ,Theoretical computer science ,Computer Networks and Communications ,Computer science ,02 engineering and technology ,Theoretical Computer Science ,Artificial Intelligence ,shortest-path ,disconnected network ,Node (computer science) ,0202 electrical engineering, electronic engineering, information engineering ,Mathematics ,Discrete mathematics ,Connected component ,routing algorithm ,business.industry ,Shortest-path tree ,020206 networking & telecommunications ,Task (computing) ,Tree (data structure) ,self-stabilization ,Hardware and Architecture ,Asynchronous communication ,Shortest path problem ,020201 artificial intelligence & image processing ,Enhanced Data Rates for GSM Evolution ,[INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC] ,Daemon ,business ,Software ,Computer network - Abstract
Many articles deal with the problem of maintaining a rooted shortest-path tree. However, after some edge deletions, some nodes can be disconnected from the connected component V r of some distinguished node r . In this case, an additional objective is to ensure the detection of the disconnection by the nodes that no longer belong to V r . We present a detailed analysis of a silent self-stabilizing algorithm. We prove that it solves this more demanding task in anonymous weighted networks with the following additional strong properties: it runs without any knowledge on the network and under the unfair daemon, that is without any assumption on the asynchronous model. Moreover, it terminates in less than 2 n + D rounds for a network of n nodes and hop-diameter D .
- Published
- 2014
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.