Search

Clothing recognition is a societally and commercially important yet extremely challenging problem due to large variations in clothing appearance, layering, style, and body shape and pose. In this paper, we tackle the clothing parsing problem using a retrieval-based approach. For a query image, we find similar styles from a large database of tagged fashion images and use these examples to recognize clothing items in the query. Our approach combines parsing from: pre-trained global clothing models, local clothing models learned on the fly from retrieved examples, and transferred parse-masks (Paper Doll item transfer) from retrieved examples. We evaluate our approach extensively and show significant improvements over previous state-of-the-art for both localization (clothing parsing given weak supervision in the form of tags) and detection (general clothing parsing). Our experimental results also indicate that the general pose estimation problem can benefit from clothing parsing. [ABSTRACT FROM AUTHOR]

Edge audio devices can reduce data bandwidth requirements by pre-processing input speech on the device before transmission to the cloud. As edge devices are required to ensure always-on operation, their stringent power constraints pose several design challenges and force IC designers to look for solutions that use low standby power. One promising bio-inspired approach is to combine the continuous-time analog filter channels with a small memory footprint deep neural network that is trained on edge tasks such as keyword spotting, thereby allowing all blocks to be embedded in an IC. This paper reviews the historical background of the continuous-time analog filter circuits that have been used as feature extractors for current edge audio devices. Starting from the interpretation of a basic biquad filter as a two-integrator-loop topology, we introduce the progression in the design of secondorder low-pass and band-pass filters ranging from OTA-based to source-follower-based architectures. We also derive and analyze the small-signal transfer function and discuss their usage in edge audio applications. [ABSTRACT FROM AUTHOR]

Graph learning is a popular approach for perfor ming machine learning on graph-structured data. It has revolutionized the machine learning ability to model graph data to address downstream tasks. Its application is wide due to the availability of graph data ranging from all types of networks to information systems. Most graph learning methods assume that the graph is static and its complete structure is known during training. This limits their applicability since they cannot be applied to problems where the underlying graph grows over time and/or new tasks emerge incrementally. Such applications require a lifelong learning approach that can learn the graph continuously and accommodate new information whilst retaining previously learned knowledge. Lifelong learning methods that enable continuous learning in regular domains like images and text cannot be directly applied to continuously evolving graph data, due to its irregular structure. As a result, graph lifelong learning is gaining attention from the research community. This survey paper provides a comprehensive overview of recent advancements in graph lifelong learning, including the categorization of existing methods, and the discussions of potential applications and open research problems. [ABSTRACT FROM AUTHOR]

The article presents information on the requirements for tutorials and teaching cases to be submitted to the journal "IEEE Transactions on Professional Communication."

In this paper, we propose a methodology for estimating the crowd speed using WiFi devices without relying on people to carry any device. Our approach not only enables speed estimation in the region where WiFi links are, but also in the adjacent possibly WiFi-free regions. More specifically, we use a pair of WiFi links in one region, whose RSSI measurements are then used to estimate the crowd speed, not only in this region, but also in adjacent WiFi-free regions. We first prove how the cross-correlation and the probability of crossing the two links implicitly carry key information about the pedestrian speeds and develop a mathematical model to relate them to pedestrian speeds. We then validate our approach with 108 experiments, in both indoor and outdoor, where up to 10 people walk in two adjacent areas, with a variety of speeds per region, showing that our framework can accurately estimate these speeds with only a pair of WiFi links in one region. For instance, the NMSE over all experiments is 0.18. We also evaluate our framework in a museum-type setting and estimate the popularity of different exhibits. We finally run experiments in an aisle in Costco, estimating key attributes of buyers’ behaviors. [ABSTRACT FROM AUTHOR]

The paper proposes a novel block-wise prediction paradigm based on Convolutional Neural Networks (CNNs) for lossless video coding. A deep neural network model which follows a multi-resolution design is employed for block-wise prediction. Several contributions are proposed to improve neural network training. A first contribution proposes a novel loss function formulation for an efficient network training based on a new approach for patch selection. Another contribution consists in replacing all HEVC-based angular intra-prediction modes with a CNN-based intra-prediction method, where each angular prediction mode is complemented by a CNN-based prediction mode using a specifically trained model. Another contribution consists in an efficient adaptation of the CNN-based intra-prediction residual for lossless video coding. Experimental results on standard test sequences show that the proposed coding system outperforms the HEVC standard with an average bitrate improvement of around 5%. To our knowledge, the paper is the first to replace all the traditional HEVC-based angular intra-prediction modes with an intra-prediction method based on modern Machine Learning techniques for lossless video coding applications. [ABSTRACT FROM AUTHOR]

Detailed routing is one of the most challenging aspects of the physical design process. Many of the violations that occur during the detailed routing stage stem from the placement of the cells. In this paper, we propose a deep learning framework to identify short violations that can occur during detailed routing from a placed netlist. One of the advantages of our technique is that by using the proposed deep learning-based predictor, global routing is no longer required as frequently and hence the total runtime for place and route can be significantly reduced. In this paper, we discuss the proposed framework and the methodology for analyzing the extracted features. The experimental results show that the average sensitivity, specificity, and accuracy of Eh?Predictor is above 90%. In addition, we show that Eh?Predictor is up to 14 times faster than NCTUgr for smaller designs and up to 96 times faster for larger designs. [ABSTRACT FROM AUTHOR]

In the last two decades, a variety of different types of multiobjective optimization problems (MOPs) have been extensively investigated in the evolutionary computation community. However, most existing evolutionary algorithms encounter difficulties in dealing with MOPs whose Pareto optimal solutions are sparse (i.e., most decision variables of the optimal solutions are zero), especially when the number of decision variables is large. Such large-scale sparse MOPs exist in a wide range of applications, for example, feature selection that aims to find a small subset of features from a large number of candidate features, or structure optimization of neural networks whose connections are sparse to alleviate overfitting. This paper proposes an evolutionary algorithm for solving large-scale sparse MOPs. The proposed algorithm suggests a new population initialization strategy and genetic operators by taking the sparse nature of the Pareto optimal solutions into consideration, to ensure the sparsity of the generated solutions. Moreover, this paper also designs a test suite to assess the performance of the proposed algorithm for large-scale sparse MOPs. The experimental results on the proposed test suite and four application examples demonstrate the superiority of the proposed algorithm over seven existing algorithms in solving large-scale sparse MOPs. [ABSTRACT FROM AUTHOR]

In virtual environments, spatial discrepancies between visual and haptic scenes negatively impact user performance and experience. This paper shows how spatial discrepancies due to pose differences can occur in a haptic augmented virtuality system with an encountered-type haptic display. To mitigate this problem, we propose visual guidance, an algorithm that dynamically manipulates the visual scene to compensate for discrepancies. The effectiveness of this algorithm was verified in a pair of studies involving a button pressing task and spatial discrepancies between ±150 mm and ±40°. Experimental results show that discrepant trials using the technique yield error rates and a number of speed peaks (representing the number of targeting movements) that are comparable to those attained in trials with zero spatial discrepancy. This result was also achieved without requiring a dedicated adaptation or training process, ensuring the algorithm can be used immediately by users. A pair of follow-up studies also indicates the algorithm has little impact on subjective ratings of simulator sickness, suggesting that sporadic use of the algorithm will not negatively affect user’s experience of a virtual environment. We believe that the visual guidance algorithm presented in this paper can be used to create more useful and compelling experiences in various haptic training applications incorporating encountered-type haptic displays. [ABSTRACT FROM AUTHOR]

Recently, vehicle type recognition in surveillance images with deep learning has received significant attention in various applications of intelligent transportation systems. However, annotating large-scale images from many surveillance images is tedious and time-consuming, which impedes its application in the real world. This paper aims to resolve this problem by reducing manual labeling in surveillance images, and then maximizing the effect of the few tagged data. Thus, a deep active learning method with a new query strategy is proposed in this paper for vehicle type recognition in surveillance images. First, the proposed method constructs a memory space using a large-scale fully labeled auxiliary dataset collected from the Internet. Subsequently, two metrics, the similarity measurement in memory space and the entropy, are used to simultaneously emphasize the diversity and uncertainty in the query strategy. Moreover, an additional label-consistent term apart from the hyper-parameters is used to adaptively adjust the combination of the two principles in active learning. The proposed method was evaluated on the Comprehensive Cars dataset. The experimental results demonstrated that the proposed method could effectively reduce the annotation cost by up to 40% in surveillance vehicle type recognition compared with the random selection method. [ABSTRACT FROM AUTHOR]

Generative adversarial networks (GANs) have been effective for learning generative models for real-world data. However, accompanied with the generative tasks becoming more and more challenging, existing GANs (GAN and its variants) tend to suffer from different training problems such as instability and mode collapse. In this paper, we propose a novel GAN framework called evolutionary GANs (E-GANs) for stable GAN training and improved generative performance. Unlike existing GANs, which employ a predefined adversarial objective function alternately training a generator and a discriminator, we evolve a population of generators to play the adversarial game with the discriminator. Different adversarial training objectives are employed as mutation operations and each individual (i.e., generator candidature) are updated based on these mutations. Then, we devise an evaluation mechanism to measure the quality and diversity of generated samples, such that only well-performing generator(s) are preserved and used for further training. In this way, E-GAN overcomes the limitations of an individual adversarial training objective and always preserves the well-performing offspring, contributing to progress in, and the success of GANs. Experiments on several datasets demonstrate that E-GAN achieves convincing generative performance and reduces the training problems inherent in existing GANs. [ABSTRACT FROM AUTHOR]

Existing nighttime vehicle detection methods use color as the primary cue for detecting vehicles. However, complex road and ambient lighting conditions, and camera configurations can influence the effectiveness of such explicit rule and threshold-based methods. In this paper, there are three main contributions. First, we present a novel method to detect vehicles during nighttime involving both learned classifiers and explicit rules, which can operate in the presence of varying ambient lighting conditions. The proposed method that is titled as Vehicle Detection using Active-learning during Nighttime (VeDANt) employs a modified form of active learning for training Adaboost classifiers with Haar-like features using gray-level input images. The hypothesis windows are then further verified using the proposed techniques involving perspective geometries and color information of the taillights. Second, VeDANt is extended to analyze the dynamics of the vehicles during nighttime by detecting three taillight activities—braking, turning left, and turning right. Third, we release three new and fully annotated Laboratory for Intelligent and Safe Automobiles-Night data sets with over 5000 frames for evaluation and benchmarking, which capture a variety of complex traffic and lighting conditions. Such comprehensively annotated and complex public data sets are a first in the area of nighttime vehicle detection. We show that VeDANt is able to detect vehicles during nighttime with over 98% accuracy and less than 1% false detections. [ABSTRACT FROM AUTHOR]

Virtual metrology (VM) predicts the actual measurement for ongoing semiconductor process. Optical emission spectroscopy (OES) data is often used to build up VM models, since it has a lot of information on process quality. However, it also includes significant redundant information, so it is important how to select only meaningful features. Deep learning (DL) techniques have been very successful in analyzing image data and it is tempting to apply those techniques to the OES data. In this paper, we propose DL configurations specific to OES data that outperform those used for image analysis. Specifically, our proposed method accounts for variable size data, chamber to chamber differences, condition drift due to accumulation, observation data drift due to an accumulation of deposition on a window, and the effects of maintenance. We evaluated our method on a real, mass-production dataset and compared our results with those obtained by using state-of-the-art image analysis DL techniques in the famous contest, ImageNet large-scale visual recognition challenge (ILSVRC). [ABSTRACT FROM AUTHOR]

In view of the great diversity and complexity of social images, it is of great significance to improve the performance of personalized recommendation by learning a user interest from large-scale social images. Deep learning, as the latest research in the field of artificial intelligence, provides a new personalized recommendation solution of social images for learning a user's interest. Moreover, social image sharing websites (such as Flickr) allow users to tag uploaded images with tags. As an important image semantic cue, effective tags not only represent the latent image information but also show personalized user interest. Therefore, a personalized recommendation method of social image is proposed by constructing a user-interest tree with deep features and tag trees in this paper. The main contributions of our paper are as follows: first, to efficiently make use of tags, a tag tree of social images is created by the re-ranked tags; second, for compactly representing the image content, deep features are learned by training the AlexNet network; third, a user-interest tree is constructed with deep features and tag trees that include the user-interest tree of social images and the user-interest tree of tags, respectively, and finally, a personalized recommendation system of social images is built based on a user-interest tree. Experiments on the NUS-WIDE dataset have shown that our method outperforms state-of-the-art methods in terms of both precision and recall of personalized recommendations. [ABSTRACT FROM AUTHOR]

This paper proposes a fully data-driven approach for PMU-based pre-fault dynamic security assessment (DSA) with incomplete data measurements. The generative adversarial network (GAN), which is an emerging unsupervised deep learning technique based on two contesting deep neural networks, is used to address the missing data. While the state-of-the-art methods for missing data are dependent on PMU observability, they are limited by the placement of PMU and network topologies. Distinguished from existing methods, the proposed approach is fully data-driven and can fill up incomplete PMU data independent on PMU observability and network topologies. Therefore, it is more generalized and extensible. Simulation results show that, under any PMU missing conditions, the proposed method can maintain a competitively high DSA accuracy with a much less computation complexity. [ABSTRACT FROM AUTHOR]

Distribution system state estimation (DSSE) is a core task for monitoring and control of distribution networks. Widely used algorithms such as Gauss–Newton perform poorly with the limited number of measurements typically available for DSSE, often require many iterations to obtain reasonable results, and sometimes fail to converge. DSSE is a non-convex problem, and working with a limited number of measurements further aggravates the situation, as indeterminacy induces multiple global (in addition to local) minima. Gauss–Newton is also known to be sensitive to initialization. Hence, the situation is far from ideal. It is therefore natural to ask if there is a smart way of initializing Gauss–Newton that will avoid these DSSE-specific pitfalls. This paper proposes using historical or simulation-derived data to train a shallow neural network to “learn to initialize,” that is, map the available measurements to a point in the neighborhood of the true latent states (network voltages), which is used to initialize Gauss–Newton. It is shown that this hybrid machine learning/optimization approach yields superior performance in terms of stability, accuracy, and runtime efficiency, compared to conventional optimization-only approaches. It is also shown that judicious design of the neural network training cost function helps to improve the overall DSSE performance. [ABSTRACT FROM AUTHOR]

With the increasing popularity of mobile phones and tablets, the explosive growth of query-by-capture applications calls for a compact representation of the query image feature. Compact descriptors for visual search (CDVS) is a recently released standard from the ISO/IEC moving pictures experts group, which achieves state-of-the-art performance in the context of image retrieval applications. However, they did not consider the matching characteristics in local space in a large-scale database, which might deteriorate the performance. In this paper, we propose a more compact representation with scale invariant feature transform (SIFT) descriptors for the visual query based on Grassmann manifold. Due to the drastic variations in image content, it is not sufficient to capture all the information using a single transform. To achieve more efficient representations, a SIFT manifold partition tree (SMPT) is initially constructed to divide the large dataset into small groups at multiple scales, which aims at capturing more discriminative information. Grassmann manifold is then applied to prune the SMPT and search for the most distinctive transforms. The experimental results demonstrate that the proposed framework achieves state-of-the-art performance on the standard benchmark CDVS dataset. [ABSTRACT FROM AUTHOR]

Horse-rider interaction analysis by wearable sensors is a promising tool for monitoring equestrian training. In this paper, a body sensor network (BSN) based equestrian motion analysis system is developed, which combines bespoke inertial measurement units (IMU) and MindWave electroencephalography (EEG) acquisition equipment. To fuse the mechanical and EEG signals collected from the system, emotional and attitude information can be obtained to analyze the interaction between the rider and horse in equestrian training. For motion data fusion, a novel method, exercise intensity extend kalman filter (EID-EKF), is proposed, which can also reconstruct the riders’ posture in different gaits by establishing a biomechanical model. The accuracy of our method is verified with the optical system Vicon to support the motion capture for four riding styles (walking, sitting troth, rising trot, canter). Finally, the emotion changes of the riders with different levels are quantified, and kinematic analysis is carried out by combining with inertial and emotional information. It is concluded from the experiment results that the estimation errors are well controlled, and motion patterns acquired according to the kinematic analysis are consistent with the actual situation. [ABSTRACT FROM AUTHOR]

This paper proposes a momentum algorithm based on Borges difference and an adaptive momentum (Adam) algorithm based on Borges difference to update parameters, which can adjust the momentum information more flexibly. The Borges difference is proposed from the definition of Borges derivative to be combined with the gradient algorithm in convolutional neural networks. The proposed momentum algorithm based on Borges difference and Adam algorithm based on Borges difference can be adjusted more flexibly in order to speed up the convergence. The parameter optimization algorithm with the Borges difference presents a better performance compared with the integer-order momentum algorithm and integer-order Adam algorithm, with the proposed nonlinear adjustment method for the parameter tuning of convolutional neural networks. By analyzing experimental results of Fashion-MNIST dataset and CIFAR-10 dataset, the optimization algorithms based on Borges difference proposed in this paper gain better effects on the optimization model compared with the corresponding ones based on the integer-order difference, and can speed up the convergence speed and recognition accuracy of the image recognition. [ABSTRACT FROM AUTHOR]

We present a method for predicting dense depth in scenarios where both a monocular camera and people in the scene are freely moving (right). Existing methods for recovering depth for dynamic, non-rigid objects from monocular video impose strong assumptions on the objects’ motion and may only recover sparse depth. In this paper, we take a data-driven approach and learn human depth priors from a new source of data: thousands of Internet videos of people imitating mannequins, i.e., freezing in diverse, natural poses, while a hand-held camera tours the scene (left). Because people are stationary, geometric constraints hold, thus training data can be generated using multi-view stereo reconstruction. At inference time, our method uses motion parallax cues from the static areas of the scenes to guide the depth prediction. We evaluate our method on real-world sequences of complex human actions captured by a moving hand-held camera, show improvement over state-of-the-art monocular depth prediction methods, and demonstrate various 3D effects produced using our predicted depth. [ABSTRACT FROM AUTHOR]

The rapid development of Internet technology has brought great convenience to our production life, and the ensuing security problems have become increasingly prominent. These problems threaten users’ privacy and pose significant security risks to the normal conduct of many aspects of society, such as politics, economy, culture, and people’s livelihood. The growth of the information transmission rate expands the scope of attacks and provides a more attack environment for intruders. Abnormal detection is an effective security protection technology that can monitor network transmission in real-time, effectively sense external attacks, and provide response decisions for relevant managers. The development of machine learning has also led to the development of abnormal traffic detection technology. The goal has been to use powerful and fast learning algorithms to deal with changing threats and respond in real-time. Most of the current abnormal detection research is based on simulation, using public and well-known datasets. On the one hand, the dataset contains high-dimensional massive data, which traditional machine learning methods cannot be processed. On the other hand, the labeled data scale is far behind the application requirements, and the dataset’s labels are all manually labeled, so the labeling cost is exceptionally high. This paper proposes a semi-supervised Double Deep Q-Network (SSDDQN)-based optimization method for network abnormal traffic detection, mainly based on Double Deep Q-Network (DDQN), a representative of Deep Reinforcement Learning algorithm. In SSDDQN, the current network first adopts the autoencoder to reconstruct the traffic features and then uses a deep neural network as a classifier. The target network first uses the unsupervised learning algorithm K-Means clustering and then uses deep neural network prediction. The experiment uses NSL-KDD and AWID datasets for training and testing and performs a comprehensive comparison with existing machine learning models. The experimental results show that SSDDQN has certain advantages in time complexity and achieved good results in various evaluation metrics. [ABSTRACT FROM AUTHOR]

Quality improvement teams engage in multiple learning opportunities during an improvement project. These learning opportunities consist of both induced opportunities, such as formal training efforts, and incidental opportunities due to surprises (i.e., events that violate expectations) encountered during problem solving activities. In this paper, we propose that these two types of team learning opportunities can result in method-driven learning (MDL) at the team level. Our data are collected using a unique longitudinal survey design spanning 31 weeks that involved 33 participants in five design for Six Sigma projects conducted in different behavioral healthcare organizations, yielding a total of 562 weekly survey responses. We examine two modes of surprises. Those occurring spontaneously (measured based on self-rated and coder-rated variables) and reflective surprises (measured as externally-induced treatments). Our results show that: 1) the proportion of surprises occurring spontaneously during problem solving activities is positively correlated with team learning (i.e., MDL); and 2) collective reflections on surprises also facilitate team learning. Furthermore, we find evidence that surprises occurring spontaneously can negatively moderate the effect of induced learning opportunities (measured as efforts expended on training) on team learning. However, we find no evidence of such a negative moderation effect for surprises from reflection. Overall, our paper develops a nuanced understanding about the direct and moderating effects of surprises on team learning in the behavioral healthcare context, thereby contributing not only to the quality management literature but also quality improvement practices in behavioral healthcare. [ABSTRACT FROM AUTHOR]

Traditional Artificial Intelligence (AI) based fault detection approaches need a large amount of data for the model learning. However, in a real-world system, it is very difficult and expensive to obtain massive labeled fault data. In addition, the working conditions of a motor are usually variable, conventional fault diagnosis models with weak generalization ability can only be used for fault detection under constant working condition. The performance of traditional AI based approaches decreases when the working condition changes. To this end, a novel deep Gaussian process (GP) kernel transfer based few-shot learning method (RNGPT) is proposed in this paper for the fault detection of electric machines. First, a deep residual network (ResNet) is used to extract the features of the raw data. Then, the encoded latent feature vector is fed into the GP with kernel transfer ability to make the motor fault detection and classification. The proposed method uses much less data than the traditional AI based method to achieve fault diagnosis under variable working condition, and does not cause an overfitting problem. Experimental results of two case studies demonstrate that the proposed RNGPT model can accurately and effectively detect motor faults with limited labeled data under different working conditions. Experimental results of RNGPT with radial basis function (RBF) kernel model on simulation data present that the fault detection accuracy of the proposed method is about 16% higher than the conventional deep learning methods, 6% higher than other few-shot learning based methods in 5-shot and 4% higher in 1-shot. Finally, experimental on a real-world dataset, the RNGPT-RBF model still has the highest fault diagnosis accuracy in 5-shot (99.39 $ \pm $ 0.09%) and 1-shot (98.55 $ \pm $ 0.16%). [ABSTRACT FROM AUTHOR]

As one of the main pollution sources in China, haze can blur the railway monitoring and threaten railway safety. In this paper, we propose an end-to-end multi-scale residual network (MSRN) which can achieve remarkable dehazing effect on railway monitoring images. The method optimizes the image dehazing algorithm in three aspects: network structure, loss function and hazy dataset. Firstly, inspired by the residual network, the paper presents a method of fusing multi-scale feature information based on the residual network, which can extract more effective information at different scales. Secondly, a combined loss function is designed to achieve better convergent results by balancing training time, training calculations, and precision. Thirdly, the paper synthesizes an outdoor dataset specifically for railway scenarios, which relies on real depth maps and various outdoor images. Extensive experimental results on both full reference image quality assessment and no reference image quality assessment of image restoration demonstrate that the proposed algorithm shows higher performance than the state-of-the-art algorithms. Moreover, the haze of railway monitoring images is removed under hazy weather, and the detection algorithm achieved higher detection accuracy on the images after dehazing. The proposed network structure, loss function, and hazy dataset are discussed and analyzed in detail to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Large-scale labeled data are generally required to train deep neural networks in order to obtain better performance in visual feature learning from images or videos for computer vision applications. To avoid extensive cost of collecting and annotating large-scale datasets, as a subset of unsupervised learning methods, self-supervised learning methods are proposed to learn general image and video features from large-scale unlabeled data without using any human-annotated labels. This paper provides an extensive review of deep learning-based self-supervised general visual feature learning methods from images or videos. First, the motivation, general pipeline, and terminologies of this field are described. Then the common deep neural network architectures that used for self-supervised learning are summarized. Next, the schema and evaluation metrics of self-supervised learning methods are reviewed followed by the commonly used datasets for images, videos, audios, and 3D data, as well as the existing self-supervised visual feature learning methods. Finally, quantitative performance comparisons of the reviewed methods on benchmark datasets are summarized and discussed for both image and video feature learning. At last, this paper is concluded and lists a set of promising future directions for self-supervised visual feature learning. [ABSTRACT FROM AUTHOR]

Inverter-based Volt/Var control (VVC) methods can effectively address issues of voltage deviations and constraint violations, as well as reduce network power losses in active distribution networks. Conventional VVC mainly relies on rule-based, mathematical, or heuristic methods. However, these methods can be inefficient or even infeasible for large-scale systems, especially when operational uncertainties are considered at both spatial and temporal scales. To this end, firstly, this paper formulates a multi-region coordinated VVC optimization problem to simultaneously minimize bus voltage deviations and network power losses. Furthermore, with the problem being formulated as a partially observable Markov Game, a multi-agent deep deterministic policy gradient (MADDPG) algorithm is modified to solve this problem. Each region, which is a sub-network with a control center, is modeled as an agent and learns to optimize inverter reactive power output setpoints through exploration in virtual environment and improved neural network training procedure. Spatial and temporal uncertainties of photovoltaic power generation and loads are modeled by stochastic programming as scenarios in the MADDPG algorithm. This algorithm can address the uncertainties to guarantee solution robustness. Thus, this paper eventually proposes a robust regionally coordinated VVC method, and numerical simulations indicate high computing efficiency and robust optimum. [ABSTRACT FROM AUTHOR]

This paper presents a distributionally robust network-constrained unit commitment (DR-NCUC) model considering AC network modeling and uncertainties of demands and renewable productions. The proposed model characterizes uncertain parameters using a data-driven ambiguity set constructed by training samples. The non-convex AC power flow equations are approximated by convex quadratic and McCormick relaxations. Since the proposed min-max-min DR-NCUC problem cannot be solved directly by available solvers, a new decomposition algorithm with proof of convergence is reported in this paper. The master problem of this algorithm is solved using both primal and dual cuts, while the max-min sub-problem is solved using the primal-dual hybrid gradient method, obviating the need for using duality theory. Also, an active set strategy is proposed to enhance the tractability of the decomposition algorithm by ignoring the subset of inactive constraints. The proposed model is applied to a 6-bus test system and the IEEE 118-bus test system under different conditions. These case studies illustrate the performance of the proposed DR-NCUC model to characterize uncertainties and the superiority of the proposed decomposition algorithm over other decomposition approaches using either primal or dual cuts. [ABSTRACT FROM AUTHOR]

Recent developments in the field of non-local attention (NLA) have led to a renewed interest in self-similarity-based single image super-resolution (SISR). Researchers usually use the NLA to explore non-local self-similarity (NSS) in SISR and achieve satisfactory reconstruction results. However, a surprising phenomenon that the reconstruction performance of the standard NLA is similar to that of the NLA with randomly selected regions prompted us to revisit NLA. In this paper, we first analyzed the attention map of the standard NLA from different perspectives and discovered that the resulting probability distribution always has full support for every local feature, which implies a statistical waste of assigning values to irrelevant non-local features, especially for SISR which needs to model long-range dependence with a large number of redundant non-local features. Based on these findings, we introduced a concise yet effective soft thresholding operation to obtain high-similarity-pass attention (HSPA), which is beneficial for generating a more compact and interpretable distribution. Furthermore, we derived some key properties of the soft thresholding operation that enable training our HSPA in an end-to-end manner. The HSPA can be integrated into existing deep SISR models as an efficient general building block. In addition, to demonstrate the effectiveness of the HSPA, we constructed a deep high-similarity-pass attention network (HSPAN) by integrating a few HSPAs in a simple backbone. Extensive experimental results demonstrate that HSPAN outperforms state-of-the-art approaches on both quantitative and qualitative evaluations. Our code and a pre-trained model were uploaded to GitHub (https://github.com/laoyangui/HSPAN) for validation. [ABSTRACT FROM AUTHOR]

How to model the effect of reflection is crucial for single image reflection removal (SIRR) task. Modern SIRR methods usually simplify the reflection formulation with the assumption of linear combination of a transmission layer and a reflection layer. However, the large variations in image content and the real-world picture-taking conditions often result in far more complex reflection. In this paper, we introduce a new screen-blur combination based on two important factors, namely the intensity and the blurriness of reflection, to better characterize the reflection formulation in SIRR. Specifically, we present Screen-blur Reflection Networks (SRNet), which executes the screen-blur formulation in its network design and adapts to the complex reflection on real scenes. Technically, SRNet consists of three components: a blended image generator, a reflection estimator and a reflection removal module. The image generator exploits the screen-blur combination to synthesize the training blended images. The reflection estimator learns the reflection layer and a blur degree that measures the level of blurriness for reflection. The reflection removal module further uses the blended image, blur degree and reflection layer to filter out the transmission layer in a cascaded manner. Superior results on three different SIRR methods are reported when generating the training data on the principle of the screen-blur combination. Moreover, extensive experiments on six datasets quantitatively and qualitatively demonstrate the efficacy of SRNet over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

With the proliferation of user-generated online videos, Multimodal Sentiment Analysis (MSA) has attracted increasing attention recently. Despite significant progress, there are still two major challenges on the way towards robust MSA: 1) inefficiency when modeling cross-modal interactions in unaligned multimodal data; and 2) vulnerability to random modality feature missing which typically occurs in realistic settings. In this paper, we propose a generic and unified framework to address them, named Efficient Multimodal Transformer with Dual-Level Feature Restoration (EMT-DLFR). Concretely, EMT employs utterance-level representations from each modality as the global multimodal context to interact with local unimodal features and mutually promote each other. It not only avoids the quadratic scaling cost of previous local-local cross-modal interaction methods but also leads to better performance. To improve model robustness in the incomplete modality setting, on the one hand, DLFR performs low-level feature reconstruction to implicitly encourage the model to learn semantic information from incomplete data. On the other hand, it innovatively regards complete and incomplete data as two different views of one sample and utilizes siamese representation learning to explicitly attract their high-level representations. Comprehensive experiments on three popular datasets demonstrate that our method achieves superior performance in both complete and incomplete modality settings. [ABSTRACT FROM AUTHOR]

The prior domain knowledge, i.e., the regularity property of continuous multiobjective optimization problems (MOPs), could be learned to guide the search for evolutionary multiobjective optimization. This paper proposes a learning-to-guide strategy (LGS) for assisting the search for multiobjective optimization algorithms in dealing with MOPs. The main idea behind LGS is to capture the regularity via learning techniques to guide the evolutionary search to generate promising offspring solutions. To achieve this, a generative model called the generative topographic mapping (GTM) is adopted to capture the manifold distribution of a population. A set of regular grid points in the latent space are mapped into the decision space within some manifold structures to guide the search for mating with some parents for offspring generation. Following this idea, three alternative LGS-based generation operators are developed and investigated, which combine the local and global information in the offspring generation. To learn the regularity more efficiently in an algorithm, the proposed LGS is embedded in an efficient evolutionary algorithm (called LGSEA). The LGSEA includes an incremental training procedure aimed at reducing the computational cost of GTM training by reusing the built GTM model. The developed algorithm is compared with some newly developed or classical learning-based algorithms on several benchmark problems. The results demonstrate the advantages of LGSEA over other approaches, showcasing its potential for solving complex MOPs. [ABSTRACT FROM AUTHOR]

Evolving fuzzy systems (EFSs) are now well developed and widely used, thanks to their ability to self-adapt both their structures and parameters online. Since the concept was first introduced two decades ago, many different types of EFSs have been successfully implemented. However, there are only very few works considering the stability of the EFSs, and these studies were limited to certain types of membership functions with specifically predefined parameters, which largely increases the complexity of the learning process. At the same time, stability analysis is of paramount importance for control applications and provides the theoretical guarantees for the convergence of the learning algorithms. In this paper, we introduce the stability proof of a class of EFSs based on data clouds, which are grounded at the AnYa type fuzzy systems and the recently introduced empirical data analytics (EDA) methodological framework. By employing data clouds, the class of EFSs of AnYa type considered in this paper avoids the traditional way of defining membership functions for each input variable in an explicit manner and its learning process is entirely data driven. The stability of the considered EFS of AnYa type is proven through the Lyapunov theory, and the proof of stability shows that the average identification error converges to a small neighborhood of zero. Although, the stability proof presented in this paper is specially elaborated for the considered EFS, it is also applicable to general EFSs. The proposed method is illustrated with Box–Jenkins gas furnace problem, one nonlinear system identification problem, Mackey–Glass time series prediction problem, eight real-world benchmark regression problems as well as a high-frequency trading prediction problem. Compared with other EFSs, the numerical examples show that the considered EFS in this paper provides guaranteed stability as well as a better approximation accuracy. [ABSTRACT FROM AUTHOR]

Electrical capacitance tomography exhibits great potentials in the visualization measurement of industrial processes, and high-precision images are of great significance for the reliability and usefulness of measurement results. In this paper, we propose a deep learning-based inversion method to ameliorate the reconstruction accuracy. With the aid of the deep learning methodology, the prior from the images reconstructed by a certain imaging technique to the true images is abstracted and stored in the deep extreme learning machine. A new cost function is constructed to encapsulate the prior from the proposed deep learning model and the domain expertise about imaging targets, and the split Bregman algorithm and the fast iterative shrinkage thresholding technique are combined into a new numerical method to effectively solve it to get the final reconstruction. The numerical and experimental results validate that the inversion method proposed in this paper reduces the reconstruction artifacts and deformations and leads to the much improvement in the imaging quality. [ABSTRACT FROM AUTHOR]

In this paper, we propose a training framework for an inverter-based memristive neuromorphic hardware. The framework, which is called PHAX, is a physical characteristics aware one relying on an ex-situ training approach. The considered neuromorphic circuit is highly energy efficient hybrid CMOS-memristive implementation of neuromorphic circuits. To solve the problem of high sensitivity of the training to the mismatches between the high-level mathematical modeling of the neurons and the corresponding physical characteristics, an approach for analytical yet accurate modeling of the memristive crossbar and neuron circuits is suggested. The approach, which is based on SPICE simulations, models the inverter-based neurons using a hyperbolic tangent function. To increase the training efficacy, the backpropagation training algorithm is modified by considering some constraints based on the physical characteristics of the memristive circuit. This modification along with the accurate back-annotation of the physical characteristics considerably improve the effectiveness of the ex-situ training method of the neuromorphic circuit. The results of this paper show an average reduction of $1805{\times }$ in the training runtime compared to that of the in-situ training approach. Furthermore, the results of applying the approach on the kernels of some applications such as image recognition, image processing, and financial analysis reveal that the designed neuromorphic circuits provide an average power saving (speed up) of $1478{\times }$ ( $5.2{\times }$ ) over the ASIC implementation in a 90-nm CMOS technology. [ABSTRACT FROM AUTHOR]

Within power generation, aging assets and an emphasis on more efficient operation of power systems and improved maintenance decision methods has led to a growing focus on asset prognostics. The main challenge facing the implementation of successful asset prognostics in power generation is the lack of available run-to-failure data. This paper proposes to overcome this issue by use of full-scope high-fidelity simulators to generate the run-to-failure data required. From this simulated failure data a similarity-based prognostic approach is developed for estimating the remaining useful life of a valve asset. Case study data is generated by initializing prebuilt industrial failure models within a 970 MW pressurized water reactor simulation. Such full-scope high-fidelity simulators are mainly operated for training purposes, allowing personnel to gain experience of standard operation as well as failures within a safe, simulated operating environment. This paper repurposes such a high-fidelity simulator to generate the type of data and affects that would be produced in the event of a fault. The fault scenario is then run multiple times to generate a library of failure events. This library of events was then split into training and test batches for building the prognostic model. Results are presented and conclusions drawn about the success of the technique and the use of high-fidelity simulators in this manner. [ABSTRACT FROM AUTHOR]

Although ultra wideband (UWB) positioning is considered as one of the most promising solutions for indoor positioning due to its high positioning accuracy, the accuracy in situations with a large number of users will be reduced because the time between two UWB position updates can be very high. To obtain a position estimate in between these updates, we can combine the UWB positioning with a different technology, e.g., an inertial measurement unit (IMU) that captures data from an accelerometer, gyroscope, and magnetometer. Previous research using the IMU outputs for location-based services employs the periodic behaviour of the accelerometer signal to count steps. However, most of these algorithms require extensive manual tuning of multiple parameters to obtain satisfactory accuracy. To overcome these practical issues, step counting algorithms using machine learning (ML) principles can be developed. In this paper, we consider accelerometer-based step counters using ML. As the performance and complexity of such algorithms depend on the features used in the training and inference phase, proper selection of the employed features is important. Therefore, in this paper, we propose a novel feature selection algorithm, where we first rank the features based on their Bhattacharyya coefficients and then systematically construct a subset of these ranked features. In this paper, we compare three ranking approaches and apply our algorithm to different ML algorithms employing an experimental set. Although the performance of the evaluated combinations slightly varies for different ML algorithms, their performance is comparable to state-of-the-art step counters, without the need to tune parameters manually. [ABSTRACT FROM PUBLISHER]

Measurement data generated by sensors on mobile devices, such as smartphones, wearables, and smart controllers, are affected by integration and real-world usage errors that can collectively be modeled as bias, linear/quadratic drift, and random noise. Motion tracking based on sensors, such as accelerometer, gyroscope, and magnetometer, is used in the field of mobile gaming control, wearable handwriting tracking, virtual reality controller, mobile phone motion tracking to control remote camera, and so on. Controller applications on mobile phone require motion sensor post processing algorithms that can work on infinite motion possibilities with instant response times and high accuracy. In this paper, we introduce a training-based algorithm that uses adaptive threshold recalibration for correcting these sensor errors identifying valid motion areas. We apply the adaptive error correction algorithm in mobile devices having accelerometer for generating linear motion coordinates. We also apply the algorithm to track rotational motion when accelerometer is fused with gyroscope and magnetometer sensors. The algorithm proposed in this paper removes motion sensor bias and drifts in 100% of the cases. The algorithm successfully identifies the duration of active motion on the motion sensor(s) data in 100% of the cases. We also compare the outputs of the proposed linear and rotational motion tracking algorithms with existing algorithms to show superior performance in terms of stability, noise tolerance, and motion tracking accuracy. [ABSTRACT FROM PUBLISHER]

Multiple antenna techniques that allow energy beamforming have been looked upon as a possible candidate for increasing the transfer efficiency between the energy transmitter (ET) and the energy receiver in wireless power transfer. This paper introduces a novel scheme that facilitates energy beamforming by utilizing received signal strength indicator (RSSI) values to estimate the channel. First, in the training stage, the ET will transmit using each beamforming vector in a codebook, which is predefined using a Cramer-Rao lower bound analysis. RSSI value corresponding to each beamforming vector is fed back to the ET, and these values are used to estimate the channel through a maximum likelihood analysis. The results that are obtained are remarkably simple, requires minimal processing, and can be easily implemented. The paper also validates the analytical results numerically, as well as experimentally, and it is shown that the proposed method achieves impressive results. [ABSTRACT FROM AUTHOR]

Unsupervised person re-identification (re-ID) remains a challenging task. While extensive research has focused on the framework design and loss function, this paper shows that sampling strategy plays an equally important role. We analyze the reasons for the performance differences between various sampling strategies under the same framework and loss function. We suggest that deteriorated over-fitting is an important factor causing poor performance, and enhancing statistical stability can rectify this problem. Inspired by that, a simple yet effective approach is proposed, termed group sampling, which gathers samples from the same class into groups. The model is thereby trained using normalized group samples, which helps alleviate the negative impact of individual samples. Group sampling updates the pipeline of pseudo-label generation by guaranteeing that samples are more efficiently classified into the correct classes. It regulates the representation learning process, enhancing statistical stability for feature representation in a progressive fashion. Extensive experiments on Market-1501, DukeMTMC-reID and MSMT17 show that group sampling achieves performance comparable to state-of-the-art methods and outperforms the current techniques under purely camera-agnostic settings. Code has been available at https://github.com/ucas-vg/GroupSampling. [ABSTRACT FROM AUTHOR]

In modern industries, data-driven fault detection and classification (FDC) systems can efficiently maintain industrial security and stability, while the security of the data-driven FDC system itself is rarely or even never considered. The security problem named adversarial vulnerability is the intrinsic of data-driven machine learning models, which will give incorrect predictions under the maliciously perturbed input data. This paper presents a work on this new security topic of the data-driven FDC systems, by 1) summarizing and comparing various recent and typical adversarial attack and defense methods for fault classifiers; 2) proposing novel attack and defense techniques for unsupervised fault detectors; 3) constructing a novel industrial adversarial security benchmark on FDC systems in the Tennessee-Eastman process (TEP) dataset; 4) exploring and discussing which attack is most potentially threatening for FDC systems and which defense technique is most applicable to mitigate attacks. The results reveal unique security properties of FDC systems, mainly including 1) for fault classifiers, black-box attack is close to the attack strength of white-box FGSM and the universal transferable attack is not significantly stronger than random noise; 2) weak adversarial training is excellent with high adversarial accuracy improvement and negligible clean accuracy decrease; 3) fault detectors are intrinsically more robust, and can be well protected by strong adversarial training. More intriguing properties and profound insights are demonstrated in the paper. This pioneering work could guide researchers and practitioners in discovering and navigating the field of FDC system adversarial robustness, outlining the research directions and open problems. [ABSTRACT FROM AUTHOR]

A novel method based on deep learning technique for the modal decomposition of the optical fields emerging from the few-mode fiber is demonstrated in this paper. By combining the advantages of principal component analysis (PCA) and Back-Propagation (BP) neural network, this scheme can reveal the exact superposition of eigenmodes. Firstly, PCA algorithm is applied to preprocess the target samples to reduce the computational complexity and extract the characteristics of the samples. Then, the mapping between the mode coefficients and the preprocessed near-field beam patterns is learned by using the BP neural network. The superiority of the proposed scheme is evaluated through simulation and experiment. The results show that the scheme can perform a complete modal decomposition within a few milliseconds, and it can still work well when the SNR level is as low as 20 dB. It is also worth noting that the method described in this work also has the advantages of short network training time, non-iterative, and low experimental equipment requirements. [ABSTRACT FROM AUTHOR]

Time series classification models have been garnering significant importance in the research community. However, not much research has been done on generating adversarial samples for these models. These adversarial samples can become a security concern. In this paper, we propose utilizing an adversarial transformation network (ATN) on a distilled model to attack various time series classification models. The proposed attack on the classification model utilizes a distilled model as a surrogate that mimics the behavior of the attacked classical time series classification models. Our proposed methodology is applied onto 1-nearest neighbor dynamic time warping (1-NN DTW) and a fully convolutional network (FCN), all of which are trained on 42 University of California Riverside (UCR) datasets. In this paper, we show both models were susceptible to attacks on all 42 datasets. When compared to Fast Gradient Sign Method, the proposed attack generates a larger faction of successful adversarial black-box attacks. A simple defense mechanism is successfully devised to reduce the fraction of successful adversarial samples. Finally, we recommend future researchers that develop time series classification models to incorporating adversarial data samples into their training data sets to improve resilience on adversarial samples. [ABSTRACT FROM AUTHOR]

In a typical new product development (NPD) project, there are many alternatives, ranging from incremental changes or re-engineering of an existing product in order to compete better in an existing market, to breakthrough thinking and the development of a new product using new technologies that satisfy the needs and expectations of customers in new markets. We tested the hypothesis that competition influences the decisions of entrepreneurs managing NPD projects. The findings presented in this paper are based on a controlled experiment conducted in several project management courses using a simulation game. The results reveal that a competitive environment in which participants compete for the same customers tends to impact the planning of NPD projects. The paper contributes to the managerial practice by offering integration of “What if?” analysis in the planning phase of NPD projects. It can assist entrepreneurs to evaluate the tradeoffs of the feasible alternatives in a noncompetitive environment, and explain how market conditions, such as competitor entry behavior, affect the project plan. A decision on the best project plan, among all the feasible project plans, is based on management strategy that takes into account the level of competition along with the uncertainty in the technology, market, and environment. The game, combined with a simulation-based training tool, used in this study, offer an opportunity to develop and to try different alternatives for the management of NPD projects without risking the consequences of project failure. [ABSTRACT FROM AUTHOR]