Search

The pulp and paper industry is critical to global industrial and economic development. Recently, India's pulp and paper industries have been facing severe competitive challenges. The challenges have impaired the environmental performance and resulted in the closure of several operations. Assessment and prediction of the performance of the Indian pulp and paper industry using various parameters is a critical task for researchers. This study proposes a framework for performance assessment and prediction based on Data Envelopment Analysis (DEA), Artificial Neural Networks, and Deep Learning (DL) to assist industry administration and decision-making. We presented a case study based on eight industries to demonstrate the methodology's applicability. This study analyses and predicts industry performance based on sample data observations over 30 years. The result suggests the DEA-DL-based efficiency prediction has an overall MSE of 0.08 compared with the actual efficiency. Furthermore, the efficiency rankings are compared between the three techniques. The results suggest that the integrated DEA-DL method is primarily accurate in most scenarios with the actual values. The findings of this study provide a comprehensive analysis of environmental performance for policymakers. [ABSTRACT FROM AUTHOR]

De-mining operations are of critical importance for humanitarian efforts and safety in conflict-affected regions. In this paper, we address the challenge of enhancing the accuracy and efficiency of mine detection systems. We present an innovative Deep Learning architecture tailored for pulse induction-based Metallic Mine Detectors (MMD), so called DL-MMD. Our methodology leverages deep neural networks to distinguish amongst nine distinct materials with an exceptional validation accuracy of 93.5%. This high level of precision enables us not only to differentiate between anti-personnel mines, without metal plates but also to detect minuscule 0.2-g vertical paper pins in both mineralized soil and non-mineralized environments. Moreover, through comparative analysis, we demonstrate a substantial 3% and 7% improvement (approx.) in accuracy performance compared to the traditional K-Nearest Neighbors and Support Vector Machine classifiers, respectively. The fusion of deep neural networks with the pulse induction-based MMD not only presents a cost-effective solution but also significantly expedites decision-making processes in de-mining operations, ultimately contributing to improved safety and effectiveness in these critical endeavors. [ABSTRACT FROM AUTHOR]

Automatic modulation classification (AMC) is an important process for future communication systems with prominent applications from spectrum management, and secure communication, to cognitive radio. The requirement for an efficient AMC classifier is due to its capability in blind modulation recognition, which is a difficult task in real scenarios where the limitations of traditional hardware and the complexity of channel impairments are involved. Therefore, this paper proposes a complete real-time AMC system based on software-defined radio and deep learning architecture. The system demodulation performance is verified through simulations and real channel impairment conditions to ensure reliability. With at most 6 times reduced number of parameters, two proposed models convolutional long short-term memory deep neural network and residual long short-term memory neural network also show a general improvement in classification accuracy compared with reference studies. The performance of these models at real-time AMC is tested with suitable processing time for practical applications. [ABSTRACT FROM AUTHOR]

The prevalence of cardiovascular disease (CVD) has surged in recent years, making it the foremost cause of mortality among humans. The Electrocardiogram (ECG), being one of the pivotal diagnostic tools for cardiovascular diseases, is increasingly gaining prominence in the field of machine learning. However, prevailing neural network models frequently disregard the spatial dimension features inherent in ECG signals. In this paper, we propose an ECG autoencoder network architecture incorporating low-rank attention (LRA-autoencoder). It is designed to capture potential spatial features of ECG signals by interpreting the signals from a spatial perspective and extracting correlations between different signal points. Additionally, the low-rank attention block (LRA-block) obtains spatial features of electrocardiogram signals through singular value decomposition, and then assigns these spatial features as weights to the electrocardiogram signals, thereby enhancing the differentiation of features among different categories. Finally, we utilize the ResNet-18 network classifier to assess the performance of the LRA-autoencoder on both the MIT-BIH Arrhythmia and PhysioNet Challenge 2017 datasets. The experimental results reveal that the proposed method demonstrates superior classification performance. The mean accuracy on the MIT-BIH Arrhythmia dataset is as high as 0.997, and the mean accuracy and F 1 -score on the PhysioNet Challenge 2017 dataset are 0.850 and 0.843. [ABSTRACT FROM AUTHOR]

A neural network model needs to be manually designed for ancient mural dynasty recognition, and this paper proposes an ancient mural dynasty recognition algorithm that is based on a neural architecture search (NAS). First, the structural edge information of mural images is extracted for use by the neural network model in recognizing mural missions. Second, an NAS algorithm that is based on contrast selection (CS) simplifies the architecture search to an incremental CS and then searches for the optimal network architecture on the mural dataset. Finally, the identified optimal network architecture is used for training and testing to complete the mural dynasty recognition task. The results show that the top accuracy of the proposed method on the mural dataset is 88.10%, the recall rate is 87.52%, and the precision rate is 87.69%. Each evaluation index used by the neural network model is superior to that of classical network models such as AlexNet and ResNet-50. Compared with NAS methods such as ASNG and MIGO, the accuracy of mural dynasty recognition is higher by an average of 4.27% when using the proposed method. The proposed method is verified on CIFAR-10, CIFAR-100, ImageNet16-120 and other datasets and achieves a good recognition accuracy in the NAS-bench-201 search space, which averages 93.26%, 70.73% and 45.34%, respectively, on the abovementioned datasets. [ABSTRACT FROM AUTHOR]

This special issue of the circuits, systems and signal processing journal focuses on recent advances and improvements in leading-edge integrated machine learning algorithms and signal processing systems. Although signal processing has been studied over several decades, the computer industry is only beginning to understand how signal processing techniques can be well integrated for the development of human-machine interfaces with the advancement of machine learning algorithms. As advances in signal processing tools and machine learning algorithms are becoming more powerful in terms of functionality and communicative capabilities, their contribution to the journal of circuits, system and signal processing is becoming more significant. [Extracted from the article]

In this paper, we present the envisioned style and scope of the new topic "Explanation Paradigms Leveraging Analytic Intuition" (ExPLAIn) with the International Journal on Software Tools for Technology Transfer (STTT). Intention behind this new topic is to (1) explicitly address all aspects and issues that arise when trying to, if possible, reveal and then confirm hidden properties of black-box systems, or (2) to enforce vital properties by embedding them into appropriate system contexts. Machine-learned systems, such as Deep Neural Networks, are particularly challenging black-box systems, and there is a wealth of formal methods for analysis and verification waiting to be adapted and applied. The selection of papers of this first Special Section of ExPLAIn, most of which were co-authored by editorial board members, is an illustrative example of the style and scope envisioned: In addition to methodological papers on verification, explanation, and their scalability, case studies, tool papers, literature reviews, and position papers are also welcome. [ABSTRACT FROM AUTHOR]

With the advent of the big data era, edge cloud computing has developed rapidly. In this era of popular digital music, various technologies have brought great convenience to online music education. But vast databases of digital music prevent educators from making specific-purpose choices. Music recommendation will be a potential development direction for online music education. In this paper, we propose a deep learning model based on multi-source information fusion for music recommendation under the scenario of edge-cloud computing. First, we use the music latent factor vector obtained by the Weighted Matrix Factorization (WMF) algorithm as the ground truth. Second, we build a neural network model to fuse multiple sources of music information, including music spectrum extracted from extra music information to predict the latent spatial features of music. Finally, we predict the user's preference for music through the inner product of the user vector and the music vector for recommendation. Experimental results on public datasets and real music data collected by edge devices demonstrate the effectiveness of the proposed method in music recommendation. [ABSTRACT FROM AUTHOR]

Traditional tool condition monitoring methods developed in an ideal environment are not universal in multiple working conditions considering different signal sources and recognition methods. This paper presents a novel tool condition monitoring approach that packages deep learning networks for accurate condition recognition with a cocktail solver library. First, the multisource signals from the machining process are collected and sequenced as the input of the cocktail solver library. The machining signals are transformed into a series of two-dimensional images by a continuous wavelet transform. In addition, ten pretrained networks with transfer learning are rapidly transferred with a finetuning operation, which contributes to a set of monitor networks. Three major processes are integrated by the cocktail solver library, which is the choosing dataset process for multi-signals, the training option process for network training parameters, and the network package process for the basic monitoring model. A Bayesian optimization method is employed to handle a tradeoff for these three processes to improve the prediction accuracy and reduce the recognition time. In the testing experiment, the milling datasets are used to train the model, and the results show that the accuracy of the model proposed in this paper can exceed 90%. The proposed method was also compared with other traditional methods to verify its effectiveness. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNN's) present some of the leading applications of Artificial Intelligence (AI) which have proven suitability on various machine-learning use cases. Forecasting demand of intermittent on-line sales is a task which needs to be carried out automatically for a large number of Stock Keeping Units (SKU's). This paper discusses the intermittent online sales and proposes an AI-based model for forecasting demand. We provide empirical evidence by utilizing data from 17 different sellers with approximately 3000 orders in total. Our findings indicate that thanks to their multi-layered learning structure, the DNN's can provide up to 35% better accuracy than the classic models such as Moving Average, Exponential Smoothing, Croston's method and ARIMA. Also, it was revealed that the time between orders' arrivals follow Exponential distribution and the order sizes also generally follow Exponential distribution. Thus, most of the time, Poisson Exponential distribution can be used for modelling intermittent sales process through online platforms. The analyses show that Poisson Exponential distribution can generate values close to real sales with less than 7% error margin with real data. [ABSTRACT FROM AUTHOR]

Hardware such as multipliers and dividers is necessary for all electronic systems. This paper explores Vedic mathematics techniques for high-speed and low-area multiplication. In the study of multiplication algorithms, various bits-width ranges of the Anurupyena sutra are used. Parallelism is employed to address challenging problems in recent studies. Various designs have been developed for the Field Programmable Gate Array (FPGA) implementation employing Very Large-Scale integration (VLSI) design approaches and parallel computing technology. Signal processing, machine learning, and reconfigurable computing research should be closely monitored as artificial intelligence develops. Multipliers and adders are key components of deep learning algorithms. The multiplier is an energy-intensive component of signal processing in Arithmetic Logic Unit (ALU), Convolutional Neural Networks (CNN), and Deep Neural Networks (DNN). For the DNN, this method introduces the Booth multiplier blocks and the carry-save multiplier in the Anurupyena architecture. Traditional multiplication methods like the array multiplier, Wallace multiplier, and Booth multiplier are contrasted with the Vedic mathematics algorithms. On a specific hardware platform, Vedic algorithms perform faster, use less power, and take up less space. Implementations were carried out using Verilog HDL and Xilinx Vivado 2019.1 on Kintex-7. The area and propagation delay were reduced compared to other multiplier architectures. [ABSTRACT FROM AUTHOR]

Water pollution is one of the most challenging environmental issues. A powerful tool for measuring the suitability of water for drinking is required. The Water Quality Index (WQI) is a widely used parameter for the assessment of water quality through mathematical formulas. In this paper, a Deep Neural Network (DNN) model is developed to forecast WQI based on parameters selected for the dry and wet seasons throughout the year. Statistical modeling and unsupervised machine learning techniques are used. These modelings include the Principal Component Analysis/Factor Analysis (PCA/FA) which is used to interpret seasonal changes and the sources of springs under study. The other modeling technique utilized in this study is the Hierarchical Cluster Analysis (HCA). The results of this study reveal that the developed DNN model has achieved a high accuracy of ***. The goodness of fit of the developed model using R-Squared (R2) is 0.98 which is deemed high. The Mean Square Error metric is close to zero. Furthermore, the PCA/FA revealed five major parameters that impact water quality which together account for 92% of the total variance of water quality in summer and 96% in winter. Moreover, results show that the average of the WQI for all springs is of poor water quality at 46.75% during the dry season and medium water quality at 55.5% during the wet season. [ABSTRACT FROM AUTHOR]

This special edition of the International Journal of Computer Vision is devoted to a selection of the best papers submitted to the 2018 British Machine Vision Conference (BMVC) held at Northumbria University, Newcastle upon Tyne, UK. Papers were assigned to Area Chairs (AC) and reviewers based on both AC and reviewer biddings on papers, CMT subject areas, and the Toronto Paper Matching System (TPMS) publication history. In " I Text2Sign: Towards Sign Language Production using Neural Machine Translation and Generative Adversarial Networks i ", Stoll et al. presented a system to automatically produce sign language videos from spoken language sentences. [Extracted from the article]

Introduction: The emergence of automated machine learning or AutoML has raised an interesting trend of no-code and low-code machine learning where most tasks in the machine learning pipeline can possibly be automated without support from human data scientists. While it sounds reasonable that we should leave repetitive trial-and-error tasks of designing complex network architectures and tuning a lot of hyperparameters to AutoML, leading research using AutoML is still scarce. Thereby, the overall purpose of this case study is to investigate the gap between current AutoML frameworks and practical machine learning development. Case description: First, this paper confirms the increasing trend of AutoML via an indirect indicator of the numbers of search results in Google trend, IEEE Xplore, and ACM Digital Library during 2012–2021. Then, the three most popular AutoML frameworks (i.e., Auto-Sklearn, AutoKeras, and Google Cloud AutoML) are inspected as AutoML's representatives; the inspection includes six comparative aspects. Based on the features available in the three AutoML frameworks investigated, our case study continues to observe recent machine learning research regarding the background of image-based machine learning. This is because the field of computer vision spans several levels of machine learning from basic to advanced and it has been one of the most popular fields in studying machine learning and artificial intelligence lately. Our study is specific to the context of image-based road health inspection systems as it has a long history in computer vision, allowing us to observe solution transitions from past to present. Discussion and evaluation: After confirming the rising numbers of AutoML search results in the three search engines, our study regarding the three AutoML representatives further reveals that there are many features that can be used to automate the development pipeline of image-based road health inspection systems. Nevertheless, we find that recent works in image-based road health inspection have not used any form of AutoML in their works. Digging into these recent works, there are two main problems that best conclude why most researchers do not use AutoML in their image-based road health inspection systems yet. Firstly, it is because AutoML's trial-and-error decision involves much extra computation compared to human-guided decisions. Secondly, using AutoML adds another layer of non-interpretability to a model. As these two problems are the major pain points in modern neural networks and deep learning, they may require years to resolve, delaying the mass adoption of AutoML in image-based road health inspection systems. Conclusions: In conclusion, although AutoML's utilization is not mainstream at this moment, we believe that the trend of AutoML will continue to grow. This is because there exists a demand for AutoML currently, and in the future, more demand for no-code or low-code machine learning development alternatives will grow together with the expansion of machine learning solutions. Nevertheless, this case study focuses on selected papers whose authors are researchers who can publish their works in academic conferences and journals. In the future, the study should continue to include observing novice users, non-programmer users, and machine learning practitioners in order to discover more insights from non-research perspectives. [ABSTRACT FROM AUTHOR]

The existing deep learning compilers are unable to perform efficient hardware performance-related graph fusion when both time and power consumption are considered. In addition, the compilers optimize the computational graph of deep neural networks (DNNs) by performing static graph transformation based on the greedy algorithm, only considering the runtime performance, and ignoring the cost of the tuning process. To solve these problems, this paper proposes a DNN computational graph optimization compiler (PCGC). Through the performance feedback at runtime, PCGC designs a computational graph fusion and splitting optimization strategy based on multilevel operator layer fusion-splitting rules. First, PCGC uses a rule-guided graph segmentation algorithm to recursively segment the computational graph into smaller subgraph to achieve an efficient and detailed search. Then, PCGC uses the cost model to receive feedback from hardware performance information, proposes cost model and operator fusion rules to synthetically guide the partial fusion and partitioning of nodes and edges of the computational graph, and generates optimal subgraphs flexibly according to different hardware to optimize the search space for partial fusion. Finally, we make the cost model converge quickly to the loss value we set by manually adjusting the parameters. Compared with other advanced compilers, PCGC optimizes the overall power consumption on an embedded GPU by an average of 130.5% when the time consumption on each hardware is not lower than the average time consumption. On domain-specific architecture, PCGC optimizes power consumption by an average of 66.5%. On FPGA, PCGC optimizes power consumption by 66.1%. In a sense, PCGC can achieve high-speed inference in specific power supply scenarios, reducing the carbon emissions of edge computing. [ABSTRACT FROM AUTHOR]

Spiking neural network (SNN) is a new generation of artificial neural networks (ANNs), which is more analogous with the brain. It has been widely considered with neural computing and brain-like intelligence. SNN is a sparse trigger event-driven model, and it has the characteristics of hardware friendliness and energy saving. SNN is more suitable for hardware implementation and rapid information processing. SNN is also a powerful method for deep learning (DL) to study brain-like computing. In this paper, the common SNN learning and training methods in the field of image classification are reviewed. In detail, we examine the SNN algorithms based on synaptic plasticity, approximate backpropagation (BP), and ANN to SNN. This paper comprehensively introduces and tracks the latest progress of SNN. On this basis, we also analyze and discuss the challenges and opportunities it faces. Finally, this paper prospects for the future development of SNN in the aspects of the biological mechanism, network training and design, computing platform, and interdisciplinary communication. This review can provide a reference for the research of SNN to promote its application in complex tasks. [ABSTRACT FROM AUTHOR]

Organs-at-risk contouring is time consuming and labour intensive. Automation by deep learning algorithms would decrease the workload of radiotherapists and technicians considerably. However, the variety of metrics used for the evaluation of deep learning algorithms make the results of many papers difficult to interpret and compare. In this paper, a qualitative evaluation is done on five established metrics to assess whether their values correlate with clinical usability. A total of 377 CT volumes with heart delineations were randomly selected for training and evaluation. A deep learning algorithm was used to predict the contours of the heart. A total of 101 CT slices from the validation set with the predicted contours were shown to three experienced radiologists. They examined each slice independently whether they would accept or adjust the prediction and if there were (small) mistakes. For each slice, the scores of this qualitative evaluation were then compared with the Sørensen-Dice coefficient (DC), the Hausdorff distance (HD), pixel-wise accuracy, sensitivity and precision. The statistical analysis of the qualitative evaluation and metrics showed a significant correlation. Of the slices with a DC over 0.96 (N = 20) or a 95% HD under 5 voxels (N = 25), no slices were rejected by the readers. Contours with lower DC or higher HD were seen in both rejected and accepted contours. Qualitative evaluation shows that it is difficult to use common quantification metrics as indicator for use in clinic. We might need to change the reporting of quantitative metrics to better reflect clinical acceptance. [ABSTRACT FROM AUTHOR]

Recently, deep neural network models for graph-structured data have been demonstrated to be influential in recommendation systems. Graph Neural Network (GNN), which can generate high-quality embeddings by capturing graph-structured information, is convenient for the recommendation. However, most existing GNN models mainly focus on the homogeneous graph. They cannot characterize heterogeneous and complex data in the recommendation system. Meanwhile, it is challenging to develop effective methods to mine the heterogeneity and latent correlations in the graph. In this paper, we adopt Heterogeneous Attributed Network (HAN), which involves different node types as well as rich node attributes, to model data in the recommendation system. Furthermore, we propose a novel graph neural network-based model to deal with HAN for Recommendation, called HANRec. In particular, we design a component connecting potential neighbors to explore the influence among neighbors and provide two different strategies with the attention mechanism to aggregate neighbors' information. The experimental results on two real-world datasets prove that HANRec outperforms other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In this paper we present a method that allows leveraging 3D electron density information to train a deep neural network pipeline to segment regions of high, medium and low electronegativity and classify substances as health hazardous or non-hazardous. We show that this can be used for use-cases such as cosmetics and food products. For this purpose, we first generate 3D electron density cubes using semiempirical molecular calculations for a custom European Chemicals Agency (ECHA) subset consisting of substances labelled as hazardous and non-hazardous for cosmetic usage. Together with their 3-class electronegativity maps we train a modified 3D-UNet with electron density cubes to segment reactive sites in molecules and classify substances with an accuracy of 78.1%. We perform the same process on a custom food dataset (CompFood) consisting of hazardous and non-hazardous substances compiled from European Food Safety Authority (EFSA) OpenFoodTox, Food and Drug Administration (FDA) Generally Recognized as Safe (GRAS) and FooDB datasets to achieve a classification accuracy of 64.1%. Our results show that 3D electron densities and particularly masked electron densities, calculated by taking a product of original electron densities and regions of high and low electronegativity can be used to classify molecules for different use-cases and thus serve not only to guide safe-by-design product development but also aid in regulatory decisions. Scientific contribution: We aim to contribute to the diverse 3D molecular representations used for training machine learning algorithms by showing that a deep learning network can be trained on 3D electron density representation of molecules. This approach has previously not been used to train machine learning models and it allows utilization of the true spatial domain of the molecule for prediction of properties such as their suitability for usage in cosmetics and food products and in future, to other molecular properties. The data and code used for training is accessible at https://github.com/s-singh-ivv/eDen-Substances. [ABSTRACT FROM AUTHOR]

This paper delves into the specialized domain of human action recognition, focusing on the Identification of Indian classical dance poses, specifically Bharatanatyam. Within the dance context, a "Karana" embodies a synchronized and harmonious movement encompassing body, hands, and feet, as defined by the Natyashastra. The essence of Karana lies in the amalgamation of nritta hasta (hand movements), sthaana (body postures), and chaari (leg movements). Although numerous, Natyashastra codifies 108 karanas, showcased in the intricate stone carvings adorning the Nataraj temples of Chidambaram, where Lord Shiva's association with these movements is depicted. Automating pose identification in Bharatanatyam poses challenges due to the vast array of variations, encompassing hand and body postures, mudras (hand gestures), facial expressions, and head gestures. To simplify this intricate task, this research employs image processing and automation techniques. The proposed methodology comprises four stages: acquisition and pre-processing of images involving skeletonization and Data Augmentation techniques, feature extraction from images, classification of dance poses using a deep learning network-based convolution neural network model (InceptionResNetV2), and visualization of 3D models through mesh creation from point clouds. The use of advanced technologies, such as the MediaPipe library for body key point detection and deep learning networks, streamlines the identification process. Data augmentation, a pivotal step, expands small datasets, enhancing the model's accuracy. The convolution neural network model showcased its effectiveness in accurately recognizing intricate dance movements, paving the way for streamlined analysis and interpretation. This innovative approach not only simplifies the identification of Bharatanatyam poses but also sets a precedent for enhancing accessibility and efficiency for practitioners and researchers in the Indian classical dance. [ABSTRACT FROM AUTHOR]

Spiking Neural Network (SNN) is one of the mainstream frameworks for brain-like computing and neuromorphic computing, which has the potential to overcome current AI challenges, for example, low-power learning dynamic processes. However, there is still a huge gap in performance between SNN and artificial neural networks (ANN) in traditional supervised learning. One solution for this problem is to propose a better spiking neuron model to improve its memory ability for temporal data. This paper proposed a leaky integrate-and-fire (LIF) neuron model with dynamic postsynaptic potential and a layer-span connecting method for SNN trained using backpropagation. The dynamic postsynaptic potential LIF model allows the neurons dynamically release neurotransmitters in an SNN model, which mimics the activity of biological neurons. The layer-span connecting method enhances the long-distance memory ability of SNN. We also first introduced a cosh-based surrogate gradient for the backpropagation training of SNNs. We compared the SNN with cosh-based surrogate gradient (CSNN), CSNN with dynamic postsynaptic potential (Dyn-CSNN), layer-span connecting CSNN (Las-CSNN), and SNN model with all the proposed methods (DlaCSNN-BP) in three image classification and one text classification datasets. The experimental results exhibited that proposed SNN methods could outperform most of the previously proposed SNNs and ANNs in the same network structure. Among them, the proposed DlaCSNN-BP got the best classification performance. This result indicates that our proposed method can effectively improve the effect of SNN in supervised learning and reduce the gap with deep learning. This work also provides more possibilities for putting SNN into practical application. [ABSTRACT FROM AUTHOR]

Floor localization is crucial for various applications such as emergency response and rescue, indoor positioning, and recommender systems. The existing floor localization systems have many drawbacks, like low accuracy, poor scalability, and high computational costs. In this paper, we first frame the problem of floor localization as one of learning node embeddings to predict the floor label of a subgraph. Then, we introduce FloorLocator, a deep learning-based method for floor localization that integrates efficient spiking neural networks with powerful graph neural networks. This approach offers high accuracy, easy scalability to new buildings, and computational efficiency. Experimental results on using several public datasets demonstrate that FloorLocator outperforms state-of-the-art methods. Notably, in building B0, FloorLocator achieved recognition accuracy of 95.9%, exceeding state-of-the-art methods by at least 10%. In building B1, it reached an accuracy of 82.1%, surpassing the latest methods by at least 4%. These results indicate FloorLocator's superiority in multi-floor building environment localization. [ABSTRACT FROM AUTHOR]

All-optical plasmonic switches (AOPSs) utilizing surface plasmon polaritons are well-suited for integration into photonic integrated circuits (PICs) and play a crucial role in advancing all-optical signal processing. The current AOPS design methods still rely on trial-and-error or empirical approaches. In contrast, recent deep learning (DL) advances have proven highly effective as computational tools, offering an alternative means to accelerate nanophotonics simulations. This paper proposes an innovative approach utilizing DL for spectrum prediction and inverse design of AOPS. The switches employ circular nonlinear plasmonic ring resonators (NPRRs) composed of interconnected metal–insulator–metal waveguides with a ring resonator. The NPRR switching performance is shown using the nonlinear Kerr effect. The forward model presented in this study demonstrates superior computational efficiency when compared to the finite-difference time-domain method. The model analyzes various structural parameters to predict transmission spectra with a distinctive dip. Inverse modeling enables the prediction of design parameters for desired transmission spectra. This model provides a rapid estimation of design parameters, offering a clear advantage over time-intensive conventional optimization approaches. The loss of prediction for both the forward and inverse models, when compared to simulations, is exceedingly low and on the order of 10−4. The results confirm the suitability of employing DL for forward and inverse design of AOPSs in PICs. [ABSTRACT FROM AUTHOR]

Across the globe, governments are developing policies and strategies to reduce carbon emissions to address climate change. Monitoring the impact of governments' carbon reduction policies can significantly enhance our ability to combat climate change and meet emissions reduction targets. One promising area in this regard is the role of artificial intelligence (AI) in carbon reduction policy and strategy monitoring. While researchers have explored applications of AI on data from various sources, including sensors, satellites, and social media, to identify areas for carbon emissions reduction, AI applications in tracking the effect of governments' carbon reduction plans have been limited. This study presents an AI framework based on long short-term memory (LSTM) and statistical process control (SPC) for the monitoring of variations in carbon emissions, using UK annual CO2 emission (per capita) data, covering a period between 1750 and 2021. This paper used LSTM to develop a surrogate model for the UK's carbon emissions characteristics and behaviours. As observed in our experiments, LSTM has better predictive abilities than ARIMA, Exponential Smoothing and feedforward artificial neural networks (ANN) in predicting CO2 emissions on a yearly prediction horizon. Using the deviation of the recorded emission data from the surrogate process, the variations and trends in these behaviours are then analysed using SPC, specifically Shewhart individual/moving range control charts. The result shows several assignable variations between the mid-1990s and 2021, which correlate with some notable UK government commitments to lower carbon emissions within this period. The framework presented in this paper can help identify periods of significant deviations from a country's normal CO2 emissions, which can potentially result from the government's carbon reduction policies or activities that can alter the amount of CO2 emissions. [ABSTRACT FROM AUTHOR]

This article explores the domain of accurate Electric Vehicle (EV) charge prediction, a crucial aspect of the energy consumption system. Predicting EV energy consumption is challenging due to the dynamic dependence and heterogeneity. Despite various approaches proposed in previous studies for intelligent charging, many models rely on limited inputs and ignore the non-linear interactivity between different time series. Moreover, to our knowledge, previous research has not considered the number of connected EVs during the charging procedure. This paper develops an attention-based recurrent convolutional neural network model (LA-RCNN) designed to forecast EV charging load using multivariate time series inputs, including meteorological data and the number of connected users. The proposed model incorporates multiplicative Luong Attention to identify temporal dependencies and correlations. Our objective is to predict the national charging load by considering the charging state and the number of plug-in EVs connected to various charging stations. Using real-world EV charging data from three Chinese cities, we demonstrate that the LA-RCNN model significantly enhances forecast accuracy compared to benchmark methods, reducing MAPE by 21.33% and RMSE by 18.73% as compared to LSTM models. These results highlight the importance of nonlinear attention-based architectures and diverse contextual data sources for effective EV load prediction. [ABSTRACT FROM AUTHOR]

Modern data mining techniques using machine learning (ML) and deep learning (DL) algorithms have been shown to excel in the regression-based task of materials property prediction using various materials representations. In an attempt to improve the predictive performance of the deep neural network model, researchers have tried to add more layers as well as develop new architectural components to create sophisticated and deep neural network models that can aid in the training process and improve the predictive ability of the final model. However, usually, these modifications require a lot of computational resources, thereby further increasing the already large model training time, which is often not feasible, thereby limiting usage for most researchers. In this paper, we study and propose a deep neural network framework for regression-based problems comprising of fully connected layers that can work with any numerical vector-based materials representations as model input. We present a novel deep regression neural network, iBRNet, with branched skip connections and multiple schedulers, which can reduce the number of parameters used to construct the model, improve the accuracy, and decrease the training time of the predictive model. We perform the model training using composition-based numerical vectors representing the elemental fractions of the respective materials and compare their performance against other traditional ML and several known DL architectures. Using multiple datasets with varying data sizes for training and testing, We show that the proposed iBRNet models outperform the state-of-the-art ML and DL models for all data sizes. We also show that the branched structure and usage of multiple schedulers lead to fewer parameters and faster model training time with better convergence than other neural networks. Scientific contribution: The combination of multiple callback functions in deep neural networks minimizes training time and maximizes accuracy in a controlled computational environment with parametric constraints for the task of materials property prediction. [ABSTRACT FROM AUTHOR]

Purpose: Mastoid abnormalities show different types of ear illnesses, however inadequacy of experts and low accuracy of diagnostic demand a new approach to detect these abnormalities and reduce human mistakes. The manual analysis of mastoid CT scans is time-consuming and labor-intensive. In this paper the first and robust deep learning-based approaches is introduced to diagnose mastoid abnormalities using a large database of CT images obtained in the clinical center with remarkable accuracy. Methods: In this paper, mastoid abnormalities are classified using the Xception based Convolutional Neural Network (CNN) model, with optimizer Adamax into five categories (Complete pneumatized, Opacification in pneumatization, Partial pneumatization, Opacification in partial pneumatization, None pneumatized). For this reason, a total of 24,800 slides of 152 patients were selected that include the mastoid from most upper to the lowest part of the middle ear cavity to complete the construction of the proposed deep neural network model. Results: The proposed model had the best accuracy of 87.80% (based on grader 1) and 88.44% (based on grader 2) on the 20th epoch and 87.70% (based on grader 1) and 87.56% (based on grader 2) on average and also significantly faster than other types of implemented architectures in terms of the computer running time (in seconds). The 99% confidence interval of the average accuracy was 0.012 which means that the true accuracy is 87.80% and 87.56% ± 1.2% that indicates the power of the model. Conclusions: The manual analysis of ear cavity CT scans is often time-consuming and prone to errors due to various inter- or intra operator variability studies. The proposed method can be used to automatically analyze the middle ear cavity to classify mastoid abnormalities, which is markedly faster than most types of models with the highest accuracy. [ABSTRACT FROM AUTHOR]

Computer Vision has been revolutionized by deep learning, and it is not surprising that the overwhelming majority of our selected papers make extensive use of neural networks. The flow of ideas between Computer Vision and Deep Learning continues to be bidirectional, and one of the selected papers defines a new learning procedure that may prove to be useful beyond computer vision (Wu and He [11]). The idea for a special issue about architectures and theories for computer vision came from the ECCV conference held in 2018. [Extracted from the article]

Distributed deep learning is a promising approach for training and deploying large and complex deep learning models. This paper presents a comprehensive workflow for deploying and optimizing the YOLACT instance segmentation model as on big data clusters. OpenVINO, a toolkit known for its high-speed data processing and ability to optimize deep learning models for deployment on a variety of devices, was used to optimize the YOLACT model. The model is then run on a big data cluster using BigDL, a distributed deep learning library for Apache Spark. BigDL provides a high-level programming interface for defining and training deep neural networks, making it suitable for large-scale deep learning applications. In distributed deep learning, input data is divided and distributed across multiple machines for parallel processing. This approach offers several advantages, including the ability to handle very large data that can be stored in a distributed manner, scalability to decrease processing time by increasing the number of workers, and fault tolerance. The proposed workflow was evaluated on virtual machines and Azure Databricks, a cloud-based platform for big data analytics. The results indicated that the workflow can scale to large datasets and deliver high performance on Azure Databricks. This study explores the benefits and challenges of using distributed deep learning on big data clusters for instance segmentation. Popular distributed deep learning frameworks are discussed, and BigDL is chosen. Overall, this study highlights the practicality of distributed deep learning for deploying and scaling sophisticated deep learning models on big data clusters. [ABSTRACT FROM AUTHOR]

In recent years, Deep Neural Networks (DNN) have emerged as a practical method for image recognition. The raw data, which contain sensitive information, are generally exploited within the training process. However, when the training process is outsourced to a third-party organization, the raw data should be desensitized before being transferred to protect sensitive information. Although masks are widely applied to hide important sensitive information, preventing inpainting masked images is critical, which may restore the sensitive information. The corresponding models should be adjusted for the masked images to reduce the degradation of the performance for recognition or classification tasks due to the desensitization of images. In this paper, we propose a mask-based image desensitization approach while supporting recognition. This approach consists of a mask generation algorithm and a model adjustment method. We propose exploiting an interpretation algorithm to maintain critical information for the recognition task in the mask generation algorithm. In addition, we propose a feature selection masknet as the model adjustment method to improve the performance based on the masked images. Extensive experimentation results based on multiple image datasets reveal significant advantages (up to 9.34% in terms of accuracy) of our approach for image desensitization while supporting recognition. [ABSTRACT FROM AUTHOR]

Deep learning models have demonstrated substantial progress in medical image segmentation. However, these models require large datasets for training, which can prove to be clinically difficult. Medical imaging datasets exhibit domain shift problems due to different imaging techniques, scanners, and data privacy issues and, conventional deep neural networks lack generalization capabilities, as their effectiveness decreases with different data distributions. This paper presents a deep learning-based attention-adaptive instance normalization style transfer technique to address the challenges encountered when segmenting blood vessels. The proposed methodology combines adaptive instance normalization style transfer with a dense extreme inception network and convolution block attention module to achieve the best observed vessel segmentation performance. A simple yet effective method is proposed, and it improves the generalization performance of deep neural networks in vascular segmentation. The network is trained on natural images and tested on medical images, thereby overcoming the need for a large dataset or labelled ground truth to train for vessel segmentation. The proposed technique uses experimental results from five distinct medical datasets to demonstrate higher cross-domain generalization capabilities than the state-of-the-art baselines available in the current literature, and the segmentation performance is compared qualitatively and quantitatively with other models. The results demonstrate the feasibility of generalizing our approach to various datasets. This approach overcomes the constraints of traditional deep learning algorithms, which require enormous volumes of medical data along with manually-labelled ground truth. The predictions by the proposed approach are based on natural image training and can be reliably used to detect and identify cardiac and retinal abnormalities without prior medical imaging information. [ABSTRACT FROM AUTHOR]

Automated image-based plant identification systems are black-boxes, failing to provide an explanation of a classification. Such explanations are seen as being essential by taxonomists and are part of the traditional procedure of plant identification. In this paper, we propose a different method by extracting explicit features from flower images that can be employed to generate explanations. We take the benefit of feature extraction derived from the taxonomic characteristics of plants, with the orchids as an example domain. Feature classifiers were developed using deep neural networks. Two different methods were studied: (1) a separate deep neural network was trained for every individual feature, and (2) a single, multi-label, deep neural network was trained, combining all features. The feature classifiers were tested in predicting 63 orchid species using naive Bayes (NB) and tree-augmented Bayesian networks (TAN). The results show that the accuracy of the feature classifiers is in the range 83-93%. By combining these features using NB and TAN the species can be predicted with an accuracy of 88.9%, which is better than a standard pre-trained deep neural-network architecture, but inferior to a deep learning architecture after fine-tuning of multiple layers. The proposed novel feature extraction method still performs well for identification and is explainable, as opposed to black-box solutions that only aim for the best performance. [ABSTRACT FROM AUTHOR]

The cost of collecting and annotating large-scale datasets is expensive, thus the valuable datasets can be considered as the intellectual property (IP) of the dataset creator. To date, all the copyright protection methods for deep learning focus on the copyright protection of the models, while there are no researches on copyright protection of the dataset. Protecting the intellectual property of dataset is a brand new topic which is very challenging. In this paper, we propose an authorization control method to actively protect the dataset from being used to train Deep Neural Network (DNN) models without authorization. To the best of our knowledge, this is the first work on IP protection for dataset. We generate feature space adversarial examples for clean images. Then, we utilize the modified Reversible Image Transformation to hide the clean images into the corresponding feature space adversarial examples to generate the protected images. For the unauthorized users, the model directly trained on the protected dataset will have poor inference accuracy. For the authorized users, the model can be trained on the recovered dataset and will have normal inference accuracy. Experimental results on CIFAR-10 and TinyImageNet datasets demonstrate the effectiveness of the proposed method. It is also demonstrated that the proposed method has an excellent transferability across different models. Moreover, the proposed method is robust to the adaptive attack. [ABSTRACT FROM AUTHOR]

As a first, the paper proposes modelling and learning of specific behaviors for dynamic obstacle avoidance in end-to-end motion planning. In the literature many end-to-end methods have been used in simulators to drive a car and to apply the learnt strategies to avoid the obstacles using the lane changing, following the vehicle as per the traffic rules, driving in-between the lane boundaries, and many more behaviors. The proposed method is designed to avoid obstacles in the scenarios where a dynamic obstacle is headed directly towards the robot from different directions. To avoid the critical encounter of the dynamic obstacles, we trained a novel deep neural network (DNN) with two specific behavioral obstacle avoidance strategies, namely "head-on collision avoidance" and "stop and move". These two strategies of obstacle avoidance come from the human behavior of obstacle avoidance. Looking at the current frame only, for a very similar visual display of the scenario, the two strategies have contrasting outputs and overall outcomes that makes learning very difficult. A random data recording over general simulations is unlikely to record the corner cases of both behaviors that rarely occur, and a behavior-specific training used in this paper intensifies the same cases for a better learning of the robot in such corner cases. We calculate the intention of the obstacle, whether it will move or not. This proposed method is compared with three state-of-the-art methods of motion planning, namely Timed-Elastic Band, Dynamic Window Approach and Nonlinear Probabilistic Velocity Obstacle. The proposed method beats all the state-of-the-art methods used for comparisons. [ABSTRACT FROM AUTHOR]

With the advent of Internet big data era, recommendation system has become a hot research topic of information selection. This paper studies the application of deep learning and distributed expression technology in e-commerce product advertising recommendation. In this paper, firstly, from the semantic level of advertising, we build a similarity network based on the theme distribution of advertising, and then build a deep learning model framework for advertising click through rate prediction. Finally, we propose an improved recommendation algorithm based on recurrent neural network and distributed expression. Aiming at the particularity of the recommendation algorithm, this paper improves the traditional recurrent neural network, and introduces a time window to control the hidden layer data transfer of the recurrent neural network. The experimental results show that the improved recurrent neural network model based on time window is superior to the traditional recurrent neural network model in the accuracy of recommendation system. The complexity of calculation is reduced and the accuracy of recommendation system is improved. [ABSTRACT FROM AUTHOR]

In recent years, the speaker-independent, single-channel speech separation problem has made significant progress with the development of deep neural networks (DNNs). However, separating the speech of each interested speaker from an environment that includes the speech of other speakers, background noise, and room reverberation remains challenging. In order to solve this problem, a speech separation method for a noisy reverberation environment is proposed. Firstly, the time-domain end-to-end network structure of a deep encoder/decoder dual-path neural network is introduced in this paper for speech separation. Secondly, to make the model not fall into local optimum during training, a loss function stretched optimal scale-invariant signal-to-noise ratio (SOSISNR) was proposed, inspired by the scale-invariant signal-to-noise ratio (SISNR). At the same time, in order to make the training more appropriate to the human auditory system, the joint loss function is extended based on short-time objective intelligibility (STOI). Thirdly, an alignment operation is proposed to reduce the influence of time delay caused by reverberation on separation performance. Combining the above methods, the subjective and objective evaluation metrics show that this study has better separation performance in complex sound field environments compared to the baseline methods. [ABSTRACT FROM AUTHOR]

Adaptive algorithms are widely used in deep learning because of their fast convergence. Among them, Adam is the most widely used algorithm. However, studies have shown that Adam's generalization ability is weak. AdaX is a variant of Adam, which introduces a novel second-order momentum, modifies the second-order moment of Adam, and has good generalization ability. However, these algorithms may fail to converge due to instability and extreme learning rates during training. In this paper, we propose a new adaptive and momental bound algorithm, called AdaXod, which characterizes of exponentially averaging the learning rate and is particularly useful for training deep neural networks. By setting an adaptively limited learning rate in the AdaX algorithm, the resultant AdaXod can effectively eliminate the problem of excessive learning rate in the later stage of neural networks training and thus results in stable training. We conduct extensive experiments on different datasets and verify the advantages of the AdaXod algorithm by comparing with other advanced adaptive optimization algorithms. AdaXod eliminates large learning rates during neural networks training and outperforms other optimizers, especially for some neural networks with complex structures, such as DenseNet. [ABSTRACT FROM AUTHOR]

Deep learning models are easily deceived by adversarial examples, and transferable attacks are crucial because of the inaccessibility of model information. Existing SOTA attack approaches tend to destroy important features of objects to generate adversarial examples. This paper proposes the split grid mask attack (SGMA), which reduces the intensity of model-specific features by split grid mask transformation, effectively highlighting the important features of the input image. Perturbing these important features can guide the development of adversarial examples in a more transferable direction. Specifically, we introduce the split grid mask transformation into the input image. Due to the vulnerability of model-specific features to image transformations, the intensity of model-specific features decreases after aggregation while the intensities of important features remain. The generated adversarial examples guided by destroying important features have excellent transferability. Extensive experimental results demonstrate the effectiveness of the proposed SGMA. Compared to the SOTA attack approaches, our method improves the black-box attack success rates by an average of 6.4% and 8.2% against the normally trained models and the defense ones respectively. [ABSTRACT FROM AUTHOR]

The research on node classification is based on node embeddings. Node classification accuracy can be improved if the embeddings of different nodes are well discriminated. With the rapid development of deep learning, researchers have proposed many graph neural network models (GNNs), such as GCN and GAT, which generally obtain node embeddings by aggregating neighborhood information. However, such methods only emphasize feature aggregation in neighborhoods and do not consider the class labels of nodes, which leads to the oversmoothing problem and weak differences in inter-class nodes. In this paper, we propose a gated graph attention network based on dual graph convolution for node embedding (GGAN-DGC). To strengthen the embedding difference of inter-class nodes, GGAN-DGC introduces a gated attention mechanism. This mechanism utilizes a supervised gated attention (GA) matrix to separate the GNN aggregation process according to the node class, so as to heterogenize the homogenous graphs. The GA matrix is obtained by the dual graph convolutional network (DGC), which can improve the receptive field of the original graph. In addition, GGAN-DGC adopts triplet loss as the global supervision function of node embedding, which can streng-then the class correlation of node embedding at the global level. Finally, based on the obtained node embedding, nodes can be classified correctly. The experimental results on five datasets confirm that our GGAN-DGC model performs better than other representative methods in node classification, especially for datasets with strong heterophily. In addition, we verify that GGAN-DGC can also perform better than other methods in graph classification experiments. [ABSTRACT FROM AUTHOR]

Finding the relationship between the chemical structure and physicochemical properties of odor molecules and olfactory perception prediction, i.e. quantitative structure-odor relationship (QSOR), remains a challenging, decades-old task. With the development of deep learning, data-driven methods such as convolutional neural networks or deep neural networks have gradually been used to predict QSOR. However, the differences between the molecular structure of different molecules are subtle and complex, the molecular feature descriptors are numerous and their interactions are complex. In this paper, we propose the Lateral Inhibition-inspired feature pyramid dynamic Convolutional Network, using the feature pyramid network as the backbone network to extract the odor molecular structure features, which can deal with multi-scale changes well. Imitating the lateral inhibition mechanism of animal olfactory, we add the lateral inhibition-inspired attention maps to the dynamic convolution, to improve the prediction accuracy of olfactory perception prediction. Besides, due to a large number of molecular feature descriptors and their complex interactions, we propose to add Attentional Factorization Mechanism to a deep neural network to obtain molecular descriptive features through weighted deep feature combination based on the attention mechanism. Our proposed olfactory perception prediction model noted as LIFMCN has achieved a state-of-the-art result and will help the product design and quality assessment in food, beverage, and fragrance industries. [ABSTRACT FROM AUTHOR]

In present times, data science become popular to support and improve decision-making process. Due to the accessibility of a wide application perspective of data streaming, class imbalance and concept drifting become crucial learning problems. The advent of deep learning (DL) models finds useful for the classification of concept drift in data streaming applications. This paper presents an effective class imbalance with concept drift detection (CIDD) using Adadelta optimizer-based deep neural networks (ADODNN), named CIDD-ADODNN model for the classification of highly imbalanced streaming data. The presented model involves four processes namely preprocessing, class imbalance handling, concept drift detection, and classification. The proposed model uses adaptive synthetic (ADASYN) technique for handling class imbalance data, which utilizes a weighted distribution for diverse minority class examples based on the level of difficulty in learning. Next, a drift detection technique called adaptive sliding window (ADWIN) is employed to detect the existence of the concept drift. Besides, ADODNN model is utilized for the classification processes. For increasing the classifier performance of the DNN model, ADO-based hyperparameter tuning process takes place to determine the optimal parameters of the DNN model. The performance of the presented model is evaluated using three streaming datasets namely intrusion detection (NSL KDDCup) dataset, Spam dataset, and Chess dataset. A detailed comparative results analysis takes place and the simulation results verified the superior performance of the presented model by obtaining a maximum accuracy of 0.9592, 0.9320, and 0.7646 on the applied KDDCup, Spam, and Chess dataset, respectively. [ABSTRACT FROM AUTHOR]

The practice of using deep learning methods in safety critical vision systems such as autonomous driving has come a long way. As vision systems supported by deep learning methods become ubiquitous, the possible security threats faced by these systems have come into greater focus. As it is with any artificial intelligence system, these deep neural vision networks are first trained on a data set of interest, once they start performing well, they are deployed to a real-world environment. In the training stage, deep learning systems are susceptible to data poisoning attacks. While deep neural networks have proved to be versatile in solving a host of challenges. These systems have complex data ecosystems especially in computer vision. In practice, the security threats when training these systems are often ignored while deploying these models in the real world. However, these threats pose significant risks to the overall reliability of the system. In this paper, we present the fundamentals of data poisoning attacks when training deep learning vision systems and discuss countermeasures against these types of attacks. In addition, we simulate the risk posed by a real-world data poisoning attack on a deep learning vision system and present a novel algorithm MOVCE--Model verification with Convolutional Neural Network and Word Embeddings which provides an effective countermeasure for maintaining the reliability of the system. The countermeasure described in this paper can be used on a wide variety of use cases where the risks posed by poisoning the training data are similar. [ABSTRACT FROM AUTHOR]

In this paper, we explore the relation between distributionally robust learning and different forms of regularization to enforce robustness of deep neural networks. In particular, starting from a concrete min-max distributionally robust problem, and using tools from optimal transport theory, we derive first-order and second-order approximations to the distributionally robust problem in terms of appropriate regularized risk minimization problems. In the context of deep ResNet models, we identify the structure of the resulting regularization problems as mean-field optimal control problems where the number and dimension of state variables are within a dimension-free factor of the dimension of the original unrobust problem. Using the Pontryagin maximum principles associated with these problems, we motivate a family of scalable algorithms for the training of robust neural networks. Our analysis recovers some results and algorithms known in the literature (in settings explained throughout the paper) and provides many other theoretical and algorithmic insights that to our knowledge are novel. In our analysis, we employ tools that we deem useful for a future analysis of more general adversarial learning problems. [ABSTRACT FROM AUTHOR]

Digital audio tampering detection can be used to verify the authenticity of digital audio. However, most current methods use standard electronic network frequency (ENF) databases for visual comparison analysis of ENF continuity of digital audio or perform feature extraction for classification by machine learning methods. ENF databases are usually tricky to obtain, visual methods have weak feature representation, and machine learning methods have more information loss in features, resulting in low detection accuracy. This paper proposes a fusion method of shallow and deep features to fully use ENF information by exploiting the complementary nature of features at different levels to more accurately describe the changes in inconsistency produced by tampering operations to raw digital audio. Firstly, the audio signal is band-pass filtered to obtain the ENF component. Then, the discrete Fourier transform (DFT) and Hilbert transform are performed to obtain the phase and instantaneous frequency of the ENF component. Secondly, the mean value of the sequence variation is used as the shallow feature; the feature matrix obtained by framing and reshaping of the ENF sequence is used as the input of the convolutional neural network; the characteristics of the fitted coefficients are obtained by curve fitting. Then, the local details of ENF are obtained from the feature matrix by the convolutional neural network, and the global information of ENF is obtained by fitting coefficient features through deep neural network (DNN). The depth features of ENF are composed of ENF global information and local information together. The shallow and deep features are fused using an attention mechanism to give greater weights to features useful for classification and suppress invalid features. Finally, the tampered audio is detected by downscaling and fitting with a DNN containing two fully connected layers, and classification is performed using a Softmax layer. The method achieves 97.03% accuracy on three classic databases: Carioca 1, Carioca 2, and New Spanish. In addition, we have achieved an accuracy of 88.31% on the newly constructed database GAUDI-DI. Experimental results show that the proposed method is superior to the state-of-the-art method. [ABSTRACT FROM AUTHOR]

In this paper, a new type of auxetic honeycomb is designed, which introduces arc walls into the concave hexagonal honeycomb cells and has higher specific energy absorption. The deformation modes and energy absorption of the designed honeycomb are analyzed by using three methods including finite element, compression experiment and machine learning methods. A fully connected two-hidden layer backpropagation deep neural network (BP-DNN) is developed to predict the energy absorptions of the honeycomb with different geometric parameters. It is found that the error of the validation set is low, and the average correlation coefficient of the validation set is 99.2%, which indicates that the neural network can obtain good predictions. The sensitivity analysis of the input parameters shows that the thickness t has the highest sensitivity, and the ratio of the length of the straight wall to the height has the lowest sensitivity to the energy. In addition, the neural network developed can also predict the mechanical properties of the honeycombs outside the parameter range of the training set and the results are consistent with that of the sensitivity analysis. The DNN provides a fast and accurate method for the energy absorption of honeycombs, which is expected to speed up the optimization and design process of honeycomb structures. [ABSTRACT FROM AUTHOR]

In this paper, we address the problem of spatially-varying illumination-aware indoor harmonization. Existing image harmonization works either focus on extracting no more than 2D information (e.g., low-level statistics or image filters) from the background image or rely on the non-linear representations of deep neural networks to adjust the foreground appearance. However, from a physical point of view, realistic image harmonization requires the perception of illumination at the foreground position in the scene (i.e., Spatially-Varying (SV) illumination), especially for indoor scenes. To solve indoor harmonization, we present a novel learning-based framework, which attempts to mimic the physical model of image formation. The proposed framework consists of a new neural harmonization architecture with four compact neural modules, which jointly learn SV illumination, shading, albedo, and rendering. In particular, a multilayer perceptron-based neural illumination field is designed to recover the illumination with finer details. Besides, we construct the first large-scale synthetic indoor harmonization benchmark dataset in which the foreground focuses on humans and is rendered and perturbed by SV illuminations. An object placement formula is also derived to ensure that the foreground object is placed in the background at a reasonable size. Extensive experiments on synthetic and real data demonstrate that our proposed approach achieves better results than prior works. [ABSTRACT FROM AUTHOR]

Recurrent neural networks (RNNs) are widely utilized in neural network research to capture spatiotemporal features in video data. However, their effectiveness heavily relies on the spatial features upon which they trained. This paper introduces innovative ensembles of features for constructing frame-wise structures by employing impactful neural network models with innovative training pipelines. These features are designed to enhance the recognition of hand gesture videos using RNN by leveraging temporal information. Recognizing hand gestures from videos is a complex task that presents considerable challenges. One notable challenge is the overlap in gesture motion, where different gesture categories exhibit similar hand poses within a single video clip. To overcome this issue, we were motivated to develop extensive and diverse features that offer a more comprehensive description of the gesture video clips, thereby mitigating recognition problems caused by images overlapping. Overall, our efforts to generate diverse features have yielded promising results in enhancing the recognition of hand gestures from videos, particularly in scenarios where overlap poses a significant challenge. We have combined the extracted features from a deep neural network trained from scratch with features obtained from various standard neural networks (Self-Organizing Map, Radial Base Function) that are trained to enhance the deep-trained features. The mutual arrangement for combining the shared features has configured new frame-wise image features. Furthermore, we have provided a performance comparison of the newly constructed frame-wise features through time-sharing to train RNN for recognition. The proposed models have been evaluated on two-hand gesture video datasets, where a preserving gesture sequence is crucial due to overlapping motions. Our work demonstrates a significant improvement in performance for both datasets. [ABSTRACT FROM AUTHOR]

We present here an artificial intelligence model based on deep learning for semantic segmentation of dermatoscopic images of skin neoplasms. The model was trained using an annotated image database. This paper discusses the challenge of increasing diagnostic accuracy using a two-level procedure based on artificial intelligence: 1) first, the object of medical interest is segmented; 2) the object is then classified. The present work is proposed as part of the first stage. Experimental studies (training set 583 images, test set 73 images) demonstrated high segmentation accuracy (Jaccard index 0.93) as compared with known methods. The approach proposed here can be used to recognize skin diseases using neural network technologies. [ABSTRACT FROM AUTHOR]

A Data Center Network (DCN) is composed of a large number of computing and storage nodes that are interconnected by well-organized switches. The Software-Defined Networking (SDN) based DCN (SD-DCN) improves resource utilization and provides virtual network access by separating the data plane and control plane of DCN. However, the routing strategies in current SD-DCN systems are based on traditional mechanisms that lack in real-time modification and are less efficient in resource utilization. To overcome these limitations, Convolutional Neural Network (CNN) deep learning model is proposed in this paper to improve the routing computation in SD-DCN, i.e., FAT-tree topology. The CNN deep learning model gives intelligent paths according to online training of traffic patterns. Moreover, the achieved network performance is compared with specific existing routing algorithms for SD-DCN. It is observed that the average network throughput is almost doubled for hot-spot traffic as compared with existing routing algorithms OSPF and FlowDCN. The experimental results show that, compared to ANN, the proposed model has increased the average network throughput by approximately 40%. Also, the proposed CNN model has outperformed the Artificial Neural Network (ANN) model in terms of average network delay and packet loss rate. Similarly, the overall bandwidth utilization is achieved by approximately 70% as compared to existing mechanisms. [ABSTRACT FROM AUTHOR]