Search

Deep convolutional neural networks (DCNNs) are currently the method of choice both for generative, as well as for discriminative learning in computer vision and machine learning. The success of DCNNs can be attributed to the careful selection of their building blocks (e.g., residual blocks, rectifiers, sophisticated normalization schemes, to mention but a few). In this paper, we propose $\Pi$ Π -Nets, a new class of function approximators based on polynomial expansions. $\Pi$ Π -Nets are polynomial neural networks, i.e., the output is a high-order polynomial of the input. The unknown parameters, which are naturally represented by high-order tensors, are estimated through a collective tensor factorization with factors sharing. We introduce three tensor decompositions that significantly reduce the number of parameters and show how they can be efficiently implemented by hierarchical neural networks. We empirically demonstrate that $\Pi$ Π -Nets are very expressive and they even produce good results without the use of non-linear activation functions in a large battery of tasks and signals, i.e., images, graphs, and audio. When used in conjunction with activation functions, $\Pi$ Π -Nets produce state-of-the-art results in three challenging tasks, i.e., image generation, face verification and 3D mesh representation learning. The source code is available at https://github.com/grigorisg9gr/polynomial_nets. [ABSTRACT FROM AUTHOR]

A general deep learning (DL) mechanism for a multiple hidden layer feed-forward neural network contains two parts, i.e., 1) an unsupervised greedy layer-wise training and 2) a supervised fine-tuning which is usually an iterative process. Although this mechanism has been demonstrated in many fields to be able to significantly improve the generalization of neural network, there is no clear evidence to show which one of the two parts plays the essential role for the generalization improvement, resulting in an argument within the DL community. Focusing on this argument, this paper proposes a new DL approach to train multilayer feed-forward neural networks. This approach uses restricted Boltzmann machine (RBM) as the layer-wise training and uses the generalized inverse of a matrix as the supervised fine-tuning. Different from the general deep training mechanism like back-propagation (BP), the proposed approach does not need to iteratively tune the weights, and therefore, has many advantages such as quick training, better generalization, and high understandability, etc. Experimentally, the proposed approach demonstrates an excellent performance in comparison with BP-based DL and the traditional training method for multilayer random weight neural networks. To a great extent, this paper demonstrates that the supervised part plays a more important role than the unsupervised part in DL, which provides some new viewpoints for exploring the essence of DL. [ABSTRACT FROM AUTHOR]

Stereo matching methods consist of matching cost computation and several post processing steps. Deep learning methods have greatly raised the accuracy of matching cost and achieved the lowest error rate on several public datasets. However, their generality capabilities are not the best due to potential overfitting, which is the common problem of supervised learning approaches. This paper proposes a convolutional neural network (CNN) based cost estimation method for computing the similarity of image patches. In consideration of accuracy and generalization capability, small size convolution kernels are chosen in the convolution layer and dropout in the decision layer is used for preventing overfitting. After obtaining stereo matching cost from the output of the CNN, several post-processing operations are adopted for disparity optimization, which includes semi-global matching in 1-D from different directions, a left-right consistency check, and the slanted plane smoothing method. The method is evaluated on KITTI 2012, KITTI 2015, and Middlebury stereo datasets and the experimental results on the KITTI benchmark demonstrate the competitive accuracy performance of the approach. Additionally, to test the generalization of the method, a series of extended crossover experiments are conducted in which the training samples and testing samples come from different datasets. The results indicate superior generalization capability of our method than other supervised learning methods. [ABSTRACT FROM AUTHOR]

In this paper, we experimentally and analytically evaluate the reliability of two state-of-the-art neural networks for linear regression and pattern recognition (multilayer perceptron and single-layer perceptron) implemented in a system-on-chip composed of a field-programmable gate array (FPGA) and a microprocessor. We have considered, for each neural network, three different activation function complexities, to evaluate how the implementation affects FPGAs reliability. As we show in this paper, higher complexity increases the exposed area but reduces the probability of one failure to impact the network output. In addition, we propose to distinguish between critical and tolerable errors in artificial neural networks. Experiments using a controlled heavy-ions beam show that, for both networks, only about 30% of the observed output errors actually affect the outputs correctness. We identify the causes of critical errors through fault injection, and found that faults in initial layers are more likely to significantly affect the output. [ABSTRACT FROM PUBLISHER]

Classification of 3D objects – the selection of a category in which each object belongs – is of great interest in the field of machine learning. Numerous researchers use deep neural networks to address this problem, altering the network architecture and representation of the 3D shape used as an input. To investigate the effectiveness of their approaches, we conduct an extensive survey of existing methods and identify common ideas by which we categorize them into a taxonomy. Second, we evaluate 11 selected classification networks on two 3D object datasets, extending the evaluation to a larger dataset on which most of the selected approaches have not been tested yet. For this, we provide a framework for converting shapes from common 3D mesh formats into formats native to each network, and for training and evaluating different classification approaches on this data. Despite being partially unable to reach the accuracies reported in the original papers, we compare the relative performance of the approaches as well as their performance when changing datasets as the only variable to provide valuable insights into performance on different kinds of data. We make our code available to simplify running training experiments with multiple neural networks with different prerequisites. [ABSTRACT FROM AUTHOR]

X-ray computed tomography (CT) using sparse projection views is a recent approach to reduce the radiation dose. However, due to the insufficient projection views, an analytic reconstruction approach using the filtered back projection (FBP) produces severe streaking artifacts. Recently, deep learning approaches using large receptive field neural networks such as U-Net have demonstrated impressive performance for sparse-view CT reconstruction. However, theoretical justification is still lacking. Inspired by the recent theory of deep convolutional framelets, the main goal of this paper is, therefore, to reveal the limitation of U-Net and propose new multi-resolution deep learning schemes. In particular, we show that the alternative U-Net variants such as dual frame and tight frame U-Nets satisfy the so-called frame condition which makes them better for effective recovery of high frequency edges in sparse-view CT. Using extensive experiments with real patient data set, we demonstrate that the new network architectures provide better reconstruction performance. [ABSTRACT FROM AUTHOR]

The efficiency of link adaptation in wireless communications relies greatly on the accuracy of channel knowledge and transmission mode selection. In this paper, a novel deep learning based link adaptation framework is proposed for the orthogonal frequency-division multiplexing (OFDM) systems with compressed-sensing-assisted index modulation, termed as OFDM-CSIM, communicating over millimeter-wave (mmWave) channels. To achieve link adaptation, a novel multi-layer sparse Bayesian learning (SBL) algorithm is proposed for accurately and instantaneously providing the required channel state information. Meanwhile, a deep neural networks (DNN)-assisted adaptive modulation algorithm is proposed to choose the best possible transmission mode to maximize the achievable throughput. Simulation results show that the proposed multi-layer SBL algorithm enables more accurate channel estimation than the conventional techniques. The DNN-based adaptive modulator is capable of achieving a higher throughput than the learning-assisted solution based on the $k$ nearest neighbor ($k$ -NN) algorithm, and also the classic average signal-to-noise ratio (SNR)-based solutions. Moreover, analysis shows that both the multi-layer SBL algorithm and the DNN-assisted adaptive modulator achieve better performance than their respective conventional counterparts while at a significantly lower computational complexity cost. [ABSTRACT FROM AUTHOR]

With the popularity of location-aware devices (e.g., smart phones), large amounts of location-based social media data such as check-ins are generated. This stimulates plenty of studies for POI recommendations by applying machine learning techniques. However, most of the existing studies focus on POI recommendations in the same city or region, and fail to recommend POIs for users when they travel to a new city. In this paper, we propose a novel deep neural network, named as ST-TransRec, for crossing-city POI recommendations. It integrates the deep neural network, transfer learning technique, and density-based resampling method into a unified framework. In this model, the deep neural network is used to capture users’ preferences for POIs and learn the embeddings of POIs. Besides, the transfer learning technique is employed to bridge the gap between cities that results from the city-dependent features. As the distributions over POIs are imbalanced, we design a density-based spatial resampling model which enables POIs to be well matched across cities. We conduct extensive experiments on two real-world datasets. The experimental results show the advantages of ST-TransRec over the state-of-the-art methods for crossing-city POI recommendations. [ABSTRACT FROM AUTHOR]

In general, an image attribute is a human-nameable visual property that has a semantic connotation. Appropriate modeling of the intrinsic contextual correlations among attributes plays a fundamental role in attribute detection. In this paper, we consider image attribute detection from the perspective of regularized deep learning. In particular, we propose a regularized deep belief network (rDBN) to perform the image attribute detection task, which is composed of two parts: 1) a detection DBN (dDBN) that models the joint distribution of images and their corresponding attributes, which acts as an attribute detector and 2) a contextual restricted Boltzmann machine that explicitly models the correlations among attributes acting as a regularizer that restraints the output detection result given by the dDBN to meet the contextual prior of attributes. Furthermore, we propose an efficient fine-tuning scheme that can further optimize the performance of the dDBN by backpropagation. Experimental results show that the proposed rDBN obtains improvements over the state-of-the-art methods for attribute detection on the benchmark data sets. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a discriminative representation for patterned fabric defect detection when only limited negative samples are available. Fabric patches are efficiently classified into defectless and defective categories by Fisher criterion-based stacked denoising autoencoders (FCSDA). First, fabric images are divided into patches of the same size, and both defective and defectless samples are utilized to train FCSDA. Second, test patches are classified through FCSDA into defective and defectless categories. Finally, the residual between the reconstructed image and defective patch is calculated, and the defect is located by thresholding. Experimental results demonstrate the effectiveness of the proposed scheme in the defect detection for periodic patterned fabric and more complex jacquard warp-knitted fabric. [ABSTRACT FROM AUTHOR]

While perception tasks such as visual object recognition and text understanding play an important role in human intelligence, subsequent tasks that involve inference, reasoning, and planning require an even higher level of intelligence. The past few years have seen major advances in many perception tasks using deep learning models. For higher-level inference, however, probabilistic graphical models with their Bayesian nature are still more powerful and flexible. To achieve integrated intelligence that involves both perception and inference, it is naturally desirable to tightly integrate deep learning and Bayesian models within a principled probabilistic framework, which we call Bayesian deep learning. In this unified framework, the perception of text or images using deep learning can boost the performance of higher-level inference and in return, the feedback from the inference process is able to enhance the perception of text or images. This paper proposes a general framework for Bayesian deep learning and reviews its recent applications on recommender systems, topic models, and control. In this paper, we also discuss the relationship and differences between Bayesian deep learning and other related topics such as the Bayesian treatment of neural networks. [ABSTRACT FROM PUBLISHER]

This paper deals with the use of emerging deep learning techniques in future wireless communication networks. It will be shown that the data-driven approaches should not replace, but rather complement, traditional design techniques based on mathematical models. Extensive motivation is given for why deep learning based on artificial neural networks will be an indispensable tool for the design and operation of future wireless communication networks, and our vision of how artificial neural networks should be integrated into the architecture of future wireless communication networks is presented. A thorough description of deep learning methodologies is provided, starting with the general machine learning paradigm, followed by a more in-depth discussion about deep learning and artificial neural networks, covering the most widely used artificial neural network architectures and their training methods. Deep learning will also be connected to other major learning frameworks, such as reinforcement learning and transfer learning. A thorough survey of the literature on deep learning for wireless communication networks is provided, followed by a detailed description of several novel case studies wherein the use of deep learning proves extremely useful for network design. For each case study, it will be shown how the use of (even approximate) mathematical models can significantly reduce the amount of live data that needs to be acquired/measured to implement the data-driven approaches. Finally, concluding remarks describe those that, in our opinion, are the major directions for future research in this field. [ABSTRACT FROM AUTHOR]

Recently, deep learning techniques have substantially boosted the performance of salient object detection in still images. However, the salient object detection in videos by using traditional handcrafted features or deep learning features is not fully investigated, probably due to the lack of sufficient manually labeled video data for saliency modeling, especially for the data-driven deep learning. This paper proposes a novel weakly supervised approach to the salient object detection in a video, which can learn a robust saliency prediction model by using very limited manually labeled data and a large amount of weakly labeled data that could be easily generated in a supervised approach. Furthermore, we propose a spatiotemporal cascade neural network architecture for saliency modeling, in which two fully convolutional networks are cascaded to evaluate the visual saliency from both spatial and temporal cues to lead the optimal video saliency prediction. The proposed approach is extensively evaluated on the widely used challenging data sets, and the experiments demonstrate that our proposed approach substantially outperforms the state-of-the-art salient object detection models. [ABSTRACT FROM AUTHOR]

Aiming at improving the classification accuracy with limited numbers of labeled pixels in polarimetric synthetic aperture radar (PolSAR) image classification task, this paper presents a graph-based semisupervised deep learning model for PolSAR image classification. It models the PolSAR image as an undirected graph, where the nodes correspond to the labeled and unlabeled pixels, and the weighted edges represent similarities between the pixels. Upon the graph, we design an energy function incorporating a semisupervision term, a convolutional neural network (CNN) term, and a pairwise smoothness term. The employed CNN extracts abstract and data-driven polarimetric features and outputs class label predictions to the graph model. The semisupervision term enforces the category label constraints on the human-labeled pixels. The pairwise smoothness term encourages class label smoothness and the alignment of class label boundaries with the image edges. Starting from an initialized class label map generated based on $K$ -Wishart distribution hypothesis or superpixel segmentation of PauliRGB images, we iteratively and alternately optimize the defined energy function until it converges. We conducted experiments on two real benchmark PolSAR images, and extensive experiments demonstrated that our approach achieved the state-of-the-art results for PolSAR image classification. [ABSTRACT FROM AUTHOR]

The key challenge of intelligent fault diagnosis is to develop features that can distinguish different categories. Because of the unique properties of mechanical data, predetermined features based on prior knowledge are usually used as inputs for fault classification. However, proper selection of features often requires expertise knowledge and becomes more difficult and time consuming when volume of data increases. In this paper, a novel deep learning network (LiftingNet) is proposed to learn features adaptively from raw mechanical data without prior knowledge. Inspired by convolutional neural network and second generation wavelet transform, the LiftingNet is constructed to classify mechanical data even though inputs contain considerable noise and randomness. The LiftingNet consists of split layer, predict layer, update layer, pooling layer, and full-connection layer. Different kernel sizes are allowed in convolutional layers to improve learning ability. As a multilayer neural network, deep features are learned from shallow ones to represent complex structures in raw data. Feasibility and effectiveness of the LiftingNet is validated by two motor bearing datasets. Results show that the proposed method could achieve layerwise feature learning and successfully classify mechanical data even with different rotating speed and under the influence of random noise. [ABSTRACT FROM PUBLISHER]

It is a feasible and promising way to utilize deep neural networks to learn and extract valuable features from synthetic aperture radar (SAR) images for SAR automatic target recognition (ATR). However, it is too difficult to effectively train the deep neural networks with limited raw SAR images. In this paper, we propose a new approach to do SAR ATR, in which a multiview deep learning framework was employed. Based on the multiview SAR ATR pattern, we first present a flexible mean to generate adequate multiview SAR data, which can guarantee a large amount of inputs for network training without needing many raw SAR images. Then, a unique deep convolutional neural network containing a parallel network topology with multiple inputs is adopted. The features of input SAR images from different views will be learned by the proposed network layer by layer; meanwhile, the learned features from the distinct views are fused in different layers progressively. Therefore, the proposed framework is able to achieve a superior recognition performance, and requires only a small number of raw SAR images for network training samples generation. Experimental results have shown the superiority of the proposed framework based on the Moving and Stationary Target Acquisition and Recognition data set. [ABSTRACT FROM AUTHOR]

In this paper, we seek to improve deep neural networks by generalizing the pooling operations that play a central role in the current architectures. We pursue a careful exploration of approaches to allow pooling to learn and to adapt to complex and variable patterns. The two primary directions lie in: (1) learning a pooling function via (two strategies of) combining of max and average pooling, and (2) learning a pooling function in the form of a tree-structured fusion of pooling filters that are themselves learned. In our experiments every generalized pooling operation we explore improves performance when used in place of average or max pooling. We experimentally demonstrate that the proposed pooling operations provide a boost in invariance properties relative to conventional pooling and set the state of the art on several widely adopted benchmark datasets. These benefits come with only a light increase in computational overhead during training (ranging from additional 5 to 15 percent in time complexity) and a very modest increase in the number of model parameters (e.g., additional 1, 9, and 27 parameters for mixed, gated, and 2-level tree pooling operators, respectively). To gain more insights about our proposed pooling methods, we also visualize the learned pooling masks and the embeddings of the internal feature responses for different pooling operations. Our proposed pooling operations are easy to implement and can be applied within various deep neural network architectures. [ABSTRACT FROM PUBLISHER]

This paper presents a simple yet effective supervised deep hash approach that constructs binary hash codes from labeled data for large-scale image search. We assume that the semantic labels are governed by several latent attributes with each attribute on or off, and classification relies on these attributes. Based on this assumption, our approach, dubbed supervised semantics-preserving deep hashing (SSDH), constructs hash functions as a latent layer in a deep network and the binary codes are learned by minimizing an objective function defined over classification error and other desirable hash codes properties. With this design, SSDH has a nice characteristic that classification and retrieval are unified in a single learning model. Moreover, SSDH performs joint learning of image representations, hash codes, and classification in a point-wised manner, and thus is scalable to large-scale datasets. SSDH is simple and can be realized by a slight enhancement of an existing deep architecture for classification; yet it is effective and outperforms other hashing approaches on several benchmarks and large datasets. Compared with state-of-the-art approaches, SSDH achieves higher retrieval accuracy, while the classification performance is not sacrificed. [ABSTRACT FROM PUBLISHER]

The paper presents a method using deep learning framework based on convolution neural network (CNN), for identification and localization of faults of transformer winding under impulse test. The results show that the proposed method outperforms the existing methods significantly. The present scheme eliminates the requirement of separate feature extraction and classification algorithms for the analysis of fault current patterns. A part of the proposed network performs feature learning and the other part classifies the features in a supervised manner. The method is computation intensive but capable of achieving very high degree of accuracy; on an average a margin of more than 7% compared to other published literature till date. [ABSTRACT FROM AUTHOR]

Deep learning based land cover classification algorithms have recently been proposed in the literature. In hyperspectral images (HSIs), they face the challenges of large dimensionality, spatial variability of spectral signatures, and scarcity of labeled data. In this paper, we propose an end-to-end deep learning architecture that extracts band specific spectral-spatial features and performs land cover classification. The architecture has fewer independent connection weights and thus requires fewer training samples. The method is found to outperform the highest reported accuracies on popular HSI data sets. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a deep metric learning (DML) approach for robust visual tracking under the particle filter framework. Unlike most existing appearance-based visual trackers, which use hand-crafted similarity metrics, our DML tracker learns a nonlinear distance metric to classify the target object and background regions using a feed-forward neural network architecture. Since there are usually large variations in visual objects caused by varying deformations, illuminations, occlusions, motions, rotations, scales, and cluttered backgrounds, conventional linear similarity metrics cannot work well in such scenarios. To address this, our proposed DML tracker first learns a set of hierarchical nonlinear transformations in the feed-forward neural network to project both the template and particles into the same feature space where the intra-class variations of positive training pairs are minimized and the interclass variations of negative training pairs are maximized simultaneously. Then, the candidate that is most similar to the template in the learned deep network is identified as the true target. Experiments on the benchmark data set including 51 challenging videos show that our DML tracker achieves a very competitive performance with the state-of-the-art trackers. [ABSTRACT FROM AUTHOR]

Predicting the millimeter wave (mmWave) beams and blockages using sub-6 GHz channels has the potential of enabling mobility and reliability in scalable mmWave systems. Prior work has focused on extracting spatial channel characteristics at the sub-6 GHz band and then use them to reduce the mmWave beam training overhead. This approach still requires beam refinement at mmWave and does not normally account for the different dielectric properties at the different bands. In this paper, we first prove that under certain conditions, there exist mapping functions that can predict the optimal mmWave beam and blockage status directly from the sub-6 GHz channel. These mapping functions, however, are hard to characterize analytically which motivates exploiting deep neural network models to learn them. For that, we prove that a large enough neural network can predict mmWave beams and blockages with success probabilities that can be made arbitrarily close to one. Then, we develop a deep learning model and empirically evaluate its beam/blockage prediction performance using a publicly available dataset. The results show that the proposed solution can predict the mmWave blockages with more than 90% success probability and can predict the optimal mmWave beams to approach the upper bounds while requiring no beam training overhead. [ABSTRACT FROM AUTHOR]

Virtual network embedding arranges virtual network services onto substrate network components. The performance of embedding algorithms determines the effectiveness and efficiency of a virtualized network, making it a critical part of the network virtualization technology. To achieve better performance, the algorithm needs to automatically detect the network status which is complicated and changes in a time-varying manner, and to dynamically provide solutions that can best fit the current network status. However, most existing algorithms fail to provide automatic embedding solutions in an acceptable running time. In this paper, we combine deep reinforcement learning with a novel neural network structure based on graph convolutional networks, and propose a new and efficient algorithm for automatic virtual network embedding. In addition, a parallel reinforcement learning framework is used in training along with a newly-designed multi-objective reward function, which has proven beneficial to the proposed algorithm for automatic embedding of virtual networks. Extensive simulation results under different scenarios show that our algorithm achieves best performance on most metrics compared with the existing state-of-the-art solutions, with upto 39.6% and 70.6% improvement on acceptance ratio and average revenue, respectively. Moreover, the results also demonstrate that the proposed solution possesses good robustness. [ABSTRACT FROM AUTHOR]

Many distributed deep learning systems have been published over the past few years, often accompanied by impressive performance claims. In practice these figures are often achieved in high performance computing (HPC) environments with fast InfiniBand network connections. For average deep learning practitioners this is usually an unrealistic scenario, since they cannot afford access to these facilities. Simple re-implementations of algorithms such as EASGD  for standard Ethernet environments often fail to replicate the scalability and performance of the original works. In this paper, we explore this particular problem domain and present MPCA SGD, a method for distributed training of deep neural networks that is specifically designed to run in low-budget environments. MPCA SGD tries to make the best possible use of available resources, and can operate well if network bandwidth is constrained. Furthermore, MPCA SGD runs on top of the popular Apache Spark  framework. Thus, it can easily be deployed in existing data centers and office environments where Spark is already used. When training large deep learning models in a gigabit Ethernet cluster, MPCA SGD achieves significantly faster convergence rates than many popular alternatives. For example, MPCA SGD can train ResNet-152  up to 5.3x faster than state-of-the-art systems like MXNet  , up to 5.3x faster than bulk-synchronous systems like SparkNet  and up to 5.3x faster than decentral asynchronous systems like EASGD . [ABSTRACT FROM AUTHOR]

This paper presents a weakly supervised regression model for the visual house appraisal problem, which aims to predict the value of a house from its photos and textual descriptions (e.g., number of bedrooms). The key idea of our approach is a multi-layer neural network, called multi-instance Deep Ranking and Regression (MiDRR) net, which jointly solves two coupled tasks: ranking and regression, in the multiple instance setting. The network is trained using weakly supervised data, which do not require intensive human annotations. We also design a set of human heuristics to promote deep features through imposing constraints over the solution space, e.g., a house with three bedrooms often has a higher value than that with only two bedrooms. While these constraints are specific to the studied problem, the developed formula can be easily generalized to the other regression applications. For test and evaluation purposes, we collect a comprehensive house image benchmark that includes 900,000 photos from 30,000 houses recently traded in the USA, and apply the proposed MiDRR net to predict house values. Extensive evaluations with comparisons demonstrate that additional usage of imagery data as well as human heuristics can significantly boost system performance and that the proposed MiDRR net clearly outperforms the alternative methods. [ABSTRACT FROM AUTHOR]

Model-based iterative reconstruction algorithms for low-dose X-ray computed tomography (CT) are computationally expensive. To address this problem, we recently proposed a deep convolutional neural network (CNN) for low-dose X-ray CT and won the second place in 2016 AAPM Low-Dose CT Grand Challenge. However, some of the textures were not fully recovered. To address this problem, here we propose a novel framelet-based denoising algorithm using wavelet residual network which synergistically combines the expressive power of deep learning and the performance guarantee from the framelet-based denoising algorithms. The new algorithms were inspired by the recent interpretation of the deep CNN as a cascaded convolution framelet signal representation. Extensive experimental results confirm that the proposed networks have significantly improved performance and preserve the detail texture of the original images. [ABSTRACT FROM AUTHOR]

Following the great success of deep convolutional neural networks (CNNs) in computer vision, this paper proposes a complex-valued CNN (CV-CNN) specifically for synthetic aperture radar (SAR) image interpretation. It utilizes both amplitude and phase information of complex SAR imagery. All elements of CNN including input-output layer, convolution layer, activation function, and pooling layer are extended to the complex domain. Moreover, a complex backpropagation algorithm based on stochastic gradient descent is derived for CV-CNN training. The proposed CV-CNN is then tested on the typical polarimetric SAR image classification task which classifies each pixel into known terrain types via supervised training. Experiments with the benchmark data sets of Flevoland and Oberpfaffenhofen show that the classification error can be further reduced if employing CV-CNN instead of conventional real-valued CNN with the same degrees of freedom. The performance of CV-CNN is comparable to that of existing state-of-the-art methods in terms of overall classification accuracy. [ABSTRACT FROM PUBLISHER]

In this modern era, traffic congestion has become a major source of severe negative economic and environmental impact for urban areas worldwide. One of the most efficient ways to mitigate traffic congestion is through future traffic prediction. The research field of traffic prediction has evolved greatly ever since its inception in the late 70s. Earlier studies mainly use classical statistical models such as ARIMA and its variants. Recently, researchers have started to focus on machine learning models because of their power and flexibility. As theoretical and technological advances emerge, we enter the era of deep neural network, which gained popularity due to its sheer prediction power which can be attributed to the complex and deep structure. Despite the popularity of deep neural network models in the field of traffic prediction, literature surveys of such methods are rare. In this work, we present an up-to-date survey of deep neural network for traffic prediction. We will provide a detailed explanation of popular deep neural network architectures commonly used in the traffic flow prediction literatures, categorize and describe the literatures themselves, present an overview of the commonalities and differences among different works, and finally provide a discussion regarding the challenges and future directions for this field. [ABSTRACT FROM AUTHOR]

Convolutional Neural Network (CNN) has gained attractions in image analytics and speech recognition in recent years. However, employing CNN for classification of graphs remains to be challenging. This paper presents the Ngram graph-block based convolutional neural network model for classification of graphs. Our Ngram deep learning framework consists of three novel components. First, we introduce the concept of $n$ -gram block to transform each raw graph object into a sequence of $n$ -gram blocks connected through overlapping regions. Second, we introduce a diagonal convolution step to extract local patterns and connectivity features hidden in these $n$ -gram blocks by performing $n$ -gram normalization. Finally, we develop deeper global patterns based on the local patterns and the ways that they respond to overlapping regions by building a $n$ -gram deep learning model using convolutional neural network. We evaluate the effectiveness of our approach by comparing it with the existing state of art methods using five real graph repositories from bioinformatics and social networks domains. Our results show that the Ngram approach outperforms existing methods with high accuracy and comparable performance. [ABSTRACT FROM PUBLISHER]

In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN [1] , which was the state-of-the-art, from $31$ to $50.3$ percent on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1 percent. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provides a global view for people to understand the deep learning object detection pipeline. [ABSTRACT FROM PUBLISHER]

Many studies on the prediction of manufacturing results using sensor signals have been conducted in the field of fault detection and classification (FDC) for semiconductor manufacturing processes. However, fault diagnosis used to find clues as to root causes remains a challenging area. In particular, process monitoring using neural networks has been employed to only a limited extent because it is a black box model, making the relationships between input data and output results difficult to interpret in actual manufacturing settings, despite its high classification performance. In this paper, we propose a convolutional neural network (CNN) model, named FDC-CNN, in which a receptive field tailored to multivariate sensor signals slides along the time axis, to extract fault features. This approach enables the association of the output of the first convolutional layer with the structural meaning of the raw data, making it possible to locate the variable and time information that represents process faults. In an experiment on a chemical vapor deposition process, the proposed method outperformed other deep learning models. [ABSTRACT FROM PUBLISHER]

The annihilating filter-based low-rank Hankel matrix approach (ALOHA) is one of the state-of-the-art compressed sensing approaches that directly interpolates the missing k-space data using low-rank Hankel matrix completion. The success of ALOHA is due to the concise signal representation in the k-space domain, thanks to the duality between structured low-rankness in the k-space domain and the image domain sparsity. Inspired by the recent mathematical discovery that links convolutional neural networks to Hankel matrix decomposition using data-driven framelet basis, here we propose a fully data-driven deep learning algorithm for k-space interpolation. Our network can be also easily applied to non-Cartesian k-space trajectories by simply adding an additional regridding layer. Extensive numerical experiments show that the proposed deep learning method consistently outperforms the existing image-domain deep learning approaches. [ABSTRACT FROM AUTHOR]

The success of machine learning algorithms generally depends on data representation, and we hypothesize that this is because different representations can entangle and hide more or less the different explanatory factors of variation behind the data. Although specific domain knowledge can be used to help design representations, learning with generic priors can also be used, and the quest for AI is motivating the design of more powerful representation-learning algorithms implementing such priors. This paper reviews recent work in the area of unsupervised feature learning and deep learning, covering advances in probabilistic models, autoencoders, manifold learning, and deep networks. This motivates longer term unanswered questions about the appropriate objectives for learning good representations, for computing representations (i.e., inference), and the geometrical connections between representation learning, density estimation, and manifold learning. [ABSTRACT FROM PUBLISHER]

Deep neural networks are now rivaling human accuracy in several pattern recognition problems. Compared to traditional classifiers, where features are handcrafted, neural networks learn increasingly complex features directly from the data. Instead of handcrafting the features, it is now the network architecture that is manually engineered. The network architecture parameters such as the number of layers or the number of filters per layer and their interconnections are essential for good performance. Even though basic design guidelines exist, designing a neural network is an iterative trial-and-error process that takes days or even weeks to perform due to the large datasets used for training. In this paper, we present DeepEyes, a Progressive Visual Analytics system that supports the design of neural networks during training. We present novel visualizations, supporting the identification of layers that learned a stable set of patterns and, therefore, are of interest for a detailed analysis. The system facilitates the identification of problems, such as superfluous filters or layers, and information that is not being captured by the network. We demonstrate the effectiveness of our system through multiple use cases, showing how a trained network can be compressed, reshaped and adapted to different problems. [ABSTRACT FROM PUBLISHER]

The emerging technology of reconfigurable intelligent surfaces (RISs) is provisioned as an enabler of smart wireless environments, offering a highly scalable, low-cost, hardware-efficient, and almost energy-neutral solution for dynamic control of the propagation of electromagnetic signals over the wireless medium, ultimately providing increased environmental intelligence for diverse operation objectives. One of the major challenges with the envisioned dense deployment of RISs in such reconfigurable radio environments is the efficient configuration of multiple metasurfaces with limited, or even the absence of, computing hardware. In this article, we consider multiuser and multi-RIS-empowered wireless systems and present a thorough survey of the online machine learning approaches for the orchestration of their various tunable components. Focusing on the sum-rate maximization as a representative design objective, we present a comprehensive problem formulation based on deep reinforcement learning (DRL). We detail the correspondences among the parameters of the wireless system and the DRL terminology, and devise generic algorithmic steps for the artificial neural network training and deployment while discussing their implementation details. Further practical considerations for multi-RIS-empowered wireless communications in the sixth-generation (6G) era are presented along with some key open research challenges. Different from the DRL-based status quo, we leverage the independence between the configuration of the system design parameters and the future states of the wireless environment, and present efficient multiarmed bandits approaches, whose resulting sum-rate performances are numerically shown to outperform random configurations, while being sufficiently close to the conventional deep $Q$ network (DQN) algorithm, but with lower implementation complexity. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs), regardless of their impressive performance, are vulnerable to attacks from adversarial inputs and, more recently, Trojans to misguide or hijack the decision of the model. We expose the existence of an intriguing class of spatially bounded, physically realizable, adversarial examples— Universal NaTuralistic adversarial paTches—we call TnTs, by exploring the super set of the spatially bounded adversarial example space and the natural input space within generative adversarial networks. Now, an adversary can arm themselves with a patch that is naturalistic, less malicious-looking, physically realizable, highly effective—achieving high attack success rates, and universal. A TnT is universal because any input image captured with a TnT in the scene will: i) misguide a network (untargeted attack); or ii) force the network to make a malicious decision (targeted attack). Interestingly, now, an adversarial patch attacker has the potential to exert a greater level of control—the ability to choose a location independent, natural-looking patch as a trigger in contrast to being constrained to noisy perturbations—an ability is thus far shown to be only possible with Trojan attack methods needing to interfere with the model building processes to embed a backdoor at the risk discovery; but, still realize a patch deployable in the physical world. Through extensive experiments on the large-scale visual classification task, ImageNet with evaluations across its entire validation set of 50,000 images, we demonstrate the realistic threat from TnTs and the robustness of the attack. We show a generalization of the attack to create patches achieving higher attack success rates than existing state-of-the-art methods. Our results show the generalizability of the attack to different visual classification tasks (CIFAR-10, GTSRB, PubFig) and multiple state-of-the-art deep neural networks such as WideResnet50, Inception-V3 and VGG-16. [ABSTRACT FROM AUTHOR]

Objective: We show that state-of-the-art deep neural networks achieve superior results in regression-based multi-class proportional myoelectric hand prosthesis control than two common baseline approaches, and we analyze the neural network mapping to explain why this is the case. Methods: Feedforward neural networks and baseline systems are trained on an offline corpus of 11 able-bodied subjects and 4 prosthesis wearers, using the $R^2$ score as metric. Analysis is performed using diverse qualitative and quantitative approaches, followed by a rigorous evaluation. Results: Our best neural networks have at least three hidden layers with at least 128 neurons per layer; smaller architectures, as used by many prior studies, perform substantially worse. The key to good performance is to both optimally regress the target movement, and to suppress spurious movements. Due to the properties of the underlying data, this is impossible to achieve with linear methods, but can be attained with high exactness using sufficiently large neural networks. Conclusion: Neural networks perform significantly better than common linear approaches in the given task, in particular when sufficiently large architectures are used. This can be explained by salient properties of the underlying data, and by theoretical and experimental analysis of the neural network mapping. Significance: To the best of our knowledge, this work is the first one in the field which not only reports that large and deep neural networks are superior to existing architectures, but also explains this result. [ABSTRACT FROM AUTHOR]

In the backpropagation algorithm, the error calculated from the output of the neural network should backpropagate the layers to update the weights of each layer, making it difficult to parallelize the training process and requiring frequent off-chip memory access. Local learning algorithms locally generate error signals which are used for weight updates, removing the need for backpropagation of error signals. However, prior works rely on large, complex auxiliary networks for reliable training, which results in large computational overhead and undermines the advantages of local learning. In this work, we propose a local learning algorithm that significantly reduces computational complexity as well as improves training performance. Our algorithm combines multiple consecutive layers into a block and performs local learning on a block-by-block basis, while dynamically changing block boundaries during training. In experiments, our approach achieves 95.68% and 79.42% test accuracy on the CIFAR-10 and CIFAR-100 datasets, respectively, using a small fully connected layer as auxiliary networks, closely matching the performance of the backpropagation algorithm. Multiply-accumulate (MAC) operations and off-chip memory access also reduce by up to 15% and 81% compared to backpropagation. [ABSTRACT FROM AUTHOR]

This article describes a novel approach to the problem of developing explainable machine learning models. We consider a deep reinforcement learning (DRL) model representing a highway path planning policy for autonomous highway driving . The model constitutes a mapping from the continuous multidimensional state space characterizing vehicle positions and velocities to a discrete set of actions in longitudinal and lateral direction. It is obtained by applying a customized version of the double deep $Q$ -network learning algorithm . The main idea is to approximate the DRL model with a set of IF–THEN rules that provide an alternative interpretable model, which is further enhanced by visualizing the rules. This concept is rationalized by the universal approximation properties of the rule-based models with fuzzy predicates. The proposed approach includes a learning engine composed of zero-order fuzzy rules, which generalize locally around the prototypes by using multivariate function models. The adjacent (in the data space) prototypes, which correspond to the same action, are further grouped and merged into the so-called MegaClouds reducing significantly the number of fuzzy rules. The input selection method is based on ranking the density of the individual inputs. Experimental results show that the specific DRL agent can be interpreted by approximating with families of rules of different granularity. The method is computationally efficient and can be potentially extended to addressing the explainability of the broader set of fully connected deep neural network models. [ABSTRACT FROM AUTHOR]

Layout hotspot detection is of great importance in the physical verification flow. Deep neural network models have been applied to hotspot detection and achieved great successes. The layouts can be viewed as binary images. The binarized neural network (BNN) can thus be suitable for the hotspot detection problem. In this article, we propose a new deep learning architecture based on BNNs to speed up the neural networks in hotspot detection. A new binarized residual neural network is carefully designed for hotspot detection. Experimental results on ICCAD 2012 and 2019 benchmarks show that our architecture outperforms previous hotspot detectors in detecting accuracy and has an 8 × speedup over the best deep learning-based solution. Since the BNN-based model is quite computationally efficient, a good tradeoff can be achieved between the efficiency and performance of the hotspot detector by applying ensemble learning approaches. Experimental results show that the ensemble models achieve better hotspot detection performance than the original with acceptable speed loss. [ABSTRACT FROM AUTHOR]

Deep learning is becoming an indispensable tool for various tasks in science and engineering. A critical step in constructing a reliable deep learning model is the selection of a loss function, which measures the discrepancy between the network prediction and the ground truth. While a variety of loss functions have been proposed in the literature, a truly optimal loss function that maximally utilizes the capacity of neural networks for deep learning-based decision-making has yet to be established. Here, we devise a generalized loss function with functional parameters determined adaptively during model training to provide a versatile framework for optimal neural network-based decision-making in small target segmentation. The method is showcased by more accurate detection and segmentation of lung and liver cancer tumors as compared with the current state-of-the-art. The proposed formalism opens new opportunities for numerous practical applications such as disease diagnosis, treatment planning, and prognosis. [ABSTRACT FROM AUTHOR]

The increasing computational cost of deep neural network models limits the applicability of intelligent applications on resource-constrained edge devices. While a number of neural network pruning methods have been proposed to compress the models, prevailing approaches focus only on parametric operators (e.g., convolution), which may miss optimization opportunities. In this article, we present a novel fusion-catalyzed pruning approach, called FuPruner, which simultaneously optimizes the parametric and nonparametric operators for accelerating neural networks. We introduce an aggressive fusion method to equivalently transform a model, which extends the optimization space of pruning and enables nonparametric operators to be pruned in a similar manner as parametric operators, and a dynamic filter pruning method is applied to decrease the computational cost of models while retaining the accuracy requirement. Moreover, FuPruner provides configurable optimization options for controlling fusion and pruning, allowing much more flexible performance-accuracy tradeoffs to be made. Evaluation with state-of-the-art residual neural networks on five representative intelligent edge platforms, Jetson TX2, Jetson Nano, Edge tensor processing unit, neural compute stick, and neural compute stick 2, demonstrates the effectiveness of our approach, which can accelerate the inference of models on CIFAR-10 and ImageNet datasets. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) have emerged as the method of choice for solving a wide range of machine learning tasks. The enormous computational demand posed by DNNs is a key challenge for computing system designers and has most commonly been addressed through the design of DNN accelerators. However, these specialized accelerators utilize large quantities of multiply-accumulate units and on-chip memory and are prohibitive in area and cost constrained systems such as wearable devices and IoT sensors. In this work, we take a complementary approach and improve the performance of DNNs on general-purpose processor (GPP) cores. We do so by exploiting a key attribute of DNNs, viz. sparsity or the prevalence of zero values. We propose Sparsity-aware Core Extensions (SparCE) - a set of low-overhead micro-architectural and ISA extensions that dynamically detect whether an operand (e.g., the result of a load instruction) is zero and subsequently skip a set of future instructions that use it. To maximize performance benefits, SparCE ensures that the instructions to be skipped are prevented from even being fetched, as squashing instructions comes with a penalty (e.g., a pipeline stall). SparCE consists of 2 key micro-architectural enhancements. First, a Sparsity Register File (SpRF) is utilized to track registers that are zero. Next, a Sparsity-Aware Skip Address (SASA) Table is used to indicate instruction sequences that can be skipped, and to specify conditions on SpRF registers that trigger instruction skipping. When an instruction is fetched, SparCE dynamically pre-identifies whether the following instruction(s) can be skipped, and if so appropriately modifies the program counter, thereby skipping the redundant instructions and improving performance. We model SparCE using the gem5 architectural simulator, and evaluate our approach on 6 state-of-the-art image-recognition DNNs in the context of both training and inference using the Caffe deep learning framework. On a scalar microprocessor, SparCE achieves 1.11×-1.96× speedups across both convolution and fully-connected layers that exhibit 10-90 percent sparsity. These speedups translate to 19-31 percent reduction in execution time at the overall application-level. We also evaluate SparCE on a 4-way SIMD ARMv8 processor using the OpenBLAS library, and demonstrate that SparCE achieves 8-15 percent reduction in the application-level execution time. [ABSTRACT FROM AUTHOR]

The cross-entropy loss commonly used in deep learning is closely related to the defining properties of optimal representations, but does not enforce some of the key properties. We show that this can be solved by adding a regularization term, which is in turn related to injecting multiplicative noise in the activations of a Deep Neural Network, a special case of which is the common practice of dropout. We show that our regularized loss function can be efficiently minimized using Information Dropout, a generalization of dropout rooted in information theoretic principles that automatically adapts to the data and can better exploit architectures of limited capacity. When the task is the reconstruction of the input, we show that our loss function yields a Variational Autoencoder as a special case, thus providing a link between representation learning, information theory and variational inference. Finally, we prove that we can promote the creation of optimal disentangled representations simply by enforcing a factorized prior, a fact that has been observed empirically in recent work. Our experiments validate the theoretical intuitions behind our method, and we find that Information Dropout achieves a comparable or better generalization performance than binary dropout, especially on smaller models, since it can automatically adapt the noise to the structure of the network, as well as to the test sample. [ABSTRACT FROM AUTHOR]