Search

Motion characteristics of human actions can be represented by the position variation of skeleton joints. Traditional approaches generally extract the spatial–temporal representation of the skeleton sequences with well-designed hand-crafted features. In this paper, in order to recognize actions according to the relative motion between the limbs and the trunk, we propose an end-to-end hierarchical RNN for skeleton-based action recognition. We divide human skeleton into five main parts in terms of the human physical structure, and then feed them to five independent subnets for local feature extraction. After the following hierarchical feature fusion and extraction from local to global, dimensions of the final temporal dynamics representations are reduced to the same number of action categories in the corresponding data set through a single-layer perceptron. In addition, the output of the perceptron is temporally accumulated as the input of a softmax layer for classification. Random scale and rotation transformations are employed to improve the robustness during training. We compare with five other deep RNN variants derived from our model in order to verify the effectiveness of the proposed network. In addition, we compare with several other methods on motion capture and Kinect data sets. Furthermore, we evaluate the robustness of our model trained with random scale and rotation transformations for a multiview problem. Experimental results demonstrate that our model achieves the state-of-the-art performance with high computational efficiency. [ABSTRACT FROM PUBLISHER]

Linear estimation of signals is often based on covariance matrices estimated from training, which can perform poorly if the training data are limited and the estimated covariance matrices are ill-conditioned. Shrinking the covariance matrix toward a scaled identity matrix can improve the robustness against the model uncertainty provided the shrinkage factor is appropriately chosen. This paper introduces several cross-validation schemes for choosing the shrinkage factors in applications where the covariance matrices are replaced with sample covariance matrices or constructed from least squares estimates of the linear model parameters. For cases where the training and out-of-training data are identically distributed, we derive leave-one-out cross-validation (LOOCV) schemes that repeatedly split the training data with respect to time to determine the optimal shrinkage factors for either model fitting or signal estimation. For cases where they are distributed differently, we develop alternative LOOCV schemes that repeatedly split the out-of-training observations with respect to space. We derive computationally efficient implementations of those schemes and provide examples to demonstrate their performance. [ABSTRACT FROM AUTHOR]

Recent deep learning models have demonstrated strong capabilities for classifying text and non-text components in natural images. They extract a high-level feature globally computed from a whole image component (patch), where the cluttered background information may dominate true text features in the deep representation. This leads to less discriminative power and poorer robustness. In this paper, we present a new system for scene text detection by proposing a novel text-attentional convolutional neural network (Text-CNN) that particularly focuses on extracting text-related regions and features from the image components. We develop a new learning mechanism to train the Text-CNN with multi-level and rich supervised information, including text region mask, character label, and binary text/non-text information. The rich supervision information enables the Text-CNN with a strong capability for discriminating ambiguous texts, and also increases its robustness against complicated background components. The training process is formulated as a multi-task learning problem, where low-level supervised information greatly facilitates the main task of text/non-text classification. In addition, a powerful low-level detector called contrast-enhancement maximally stable extremal regions (MSERs) is developed, which extends the widely used MSERs by enhancing intensity contrast between text patterns and background. This allows it to detect highly challenging text patterns, resulting in a higher recall. Our approach achieved promising results on the ICDAR 2013 data set, with an F-measure of 0.82, substantially improving the state-of-the-art results. [ABSTRACT FROM AUTHOR]

Design space exploration and sensitivity analysis for electrical performance of high-speed serial links is a critical and challenging task for a robust, cost-efficient, and signal-integrity-compliant channel design. The generation of time-domain (TD) metrics like eye height and eye width at higher bit error rates requires longer bit sequences in TD circuit simulation, which is compute time intensive. Intelligent techniques to identify smaller design sets that cover the design space optimally may provide incorrect sensitivity analysis. This paper explores learning-based modeling techniques that rapidly map relevant frequency-domain metrics like differential insertion loss and total cross talk, in the presence of equalization, to TD metrics like eye height and eye width, thus facilitating a full-factorial design space sweep. Numerical results performed with multilayer-perceptron-based artificial neural network as well as least-squares support vector machine (LS-SVM) on Serial ATA 3.0 and Peripheral Component Interconnect Express Gen3 channels generate an average error of less than 2%. [ABSTRACT FROM PUBLISHER]

To improve the efficiency of big data feature learning, the paper proposes a privacy preserving deep computation model by offloading the expensive operations to the cloud. Privacy concerns become evident because there are a large number of private data by various applications in the smart city, such as sensitive data of governments or proprietary information of enterprises. To protect the private data, the proposed model uses the BGV encryption scheme to encrypt the private data and employs cloud servers to perform the high-order back-propagation algorithm on the encrypted data efficiently for deep computation model training. Furthermore, the proposed scheme approximates the Sigmoid function as a polynomial function to support the secure computation of the activation function with the BGV encryption. In our scheme, only the encryption operations and the decryption operations are performed by the client while all the computation tasks are performed on the cloud. Experimental results show that our scheme is improved by approximately 2.5 times in the training efficiency compared to the conventional deep computation model without disclosing the private data using the cloud computing including ten nodes. More importantly, our scheme is highly scalable by employing more cloud servers, which is particularly suitable for big data. [ABSTRACT FROM AUTHOR]

Solving the problem of matching people across non-overlapping multi-camera views, known as person re-identification (re-id), has received increasing interests in computer vision. In a real-world application scenario, a watch-list (gallery set) of a handful of known target people are provided with very few (in many cases only a single) image(s) (shots) per target. Existing re-id methods are largely unsuitable to address this open-world re-id challenge because they are designed for (1) a closed-world scenario where the gallery and probe sets are assumed to contain exactly the same people, (2) person-wise identification whereby the model attempts to verify exhaustively against each individual in the gallery set, and (3) learning a matching model using multi-shots. In this paper, a novel transfer local relative distance comparison (t-LRDC) model is formulated to address the open-world person re-identification problem by one-shot group-based verification. The model is designed to mine and transfer useful information from a labelled open-world non-target dataset. Extensive experiments demonstrate that the proposed approach outperforms both non-transfer learning and existing transfer learning based re-id methods. [ABSTRACT FROM PUBLISHER]

The standard Gaussian Process regression (GP) is usually formulated under stationary hypotheses: The noise power is considered constant throughout the input space and the covariance of the prior distribution is typically modeled as depending only on the difference between input samples. These assumptions can be too restrictive and unrealistic for many real-world problems. Although nonstationarity can be achieved using specific covariance functions, they require a prior knowledge of the kind of nonstationarity, not available for most applications. In this paper we propose to use the Laplace approximation to make inference in a divisive GP model to perform nonstationary regression, including heteroscedastic noise cases. The log-concavity of the likelihood ensures a unimodal posterior and makes that the Laplace approximation converges to a unique maximum. The characteristics of the likelihood also allow to obtain accurate posterior approximations when compared to the Expectation Propagation (EP) approximations and the asymptotically exact posterior provided by a Markov Chain Monte Carlo implementation with Elliptical Slice Sampling (ESS), but at a reduced computational load with respect to both, EP and ESS. [ABSTRACT FROM PUBLISHER]

Localizing objects of interest in images when provided with only image-level labels is a challenging visual recognition task. Previous efforts have required carefully designed features and have difficulty in handling images with cluttered backgrounds. Up-scaling to large datasets also poses a challenge to applying these methods to real applications. In this paper, we propose an efficient and effective learning framework called MILinear, which is able to learn an object localization model from large-scale data without using bounding box annotations. We integrate rich general prior knowledge into a learning model using a large pre-trained convolutional network. Moreover, to reduce ambiguity in positive images, we present a bag-splitting algorithm that iteratively generates new negative bags from positive ones. We evaluate the proposed approach on the challenging Pascal VOC 2007 dataset, and our method outperforms other state-of-the-art methods by a large margin; some results are even comparable to fully supervised models trained with bounding box annotations. To further demonstrate scalability, we also present detection results on the ILSVRC 2013 detection dataset, and our method outperforms supervised deformable part-based model without using box annotations. [ABSTRACT FROM PUBLISHER]

Discriminative methods (e.g., kernel regression, SVM) have been extensively used to solve problems such as object recognition, image alignment and pose estimation from images. These methods typically map image features ( \mathbf X ) to continuous (e.g., pose) or discrete (e.g., object category) values. A major drawback of existing discriminative methods is that samples are directly projected onto a subspace and hence fail to account for outliers common in realistic training sets due to occlusion, specular reflections or noise. It is important to notice that existing discriminative approaches assume the input variables \mathbf X to be noise free. Thus, discriminative methods experience significant performance degradation when gross outliers are present. Despite its obvious importance, the problem of robust discriminative learning has been relatively unexplored in computer vision. This paper develops the theory of robust regression (RR) and presents an effective convex approach that uses recent advances on rank minimization. The framework applies to a variety of problems in computer vision including robust linear discriminant analysis, regression with missing data, and multi-label classification. Several synthetic and real examples with applications to head pose estimation from images, image and video classification and facial attribute classification with missing data are used to illustrate the benefits of RR. [ABSTRACT FROM PUBLISHER]

Accurate forecasts of electrical substations are mandatory for the efficiency of the Advanced Distribution Automation functions in distribution systems. The paper describes the design of a class of machine-learning models, namely neural networks, for the load forecasts of medium-voltage/low-voltage substations. We focus on the methodology of neural network model design in order to obtain a model that has the best achievable predictive ability given the available data. Variable selection and model selection are applied to electrical load forecasts to ensure an optimal generalization capacity of the neural network model. Real measurements collected in French distribution systems are used to validate our study. The results show that the neural network-based models outperform the time series models and that the design methodology guarantees the best generalization ability of the neural network model for the load forecasting purpose based on the same data. [ABSTRACT FROM AUTHOR]

Rare class problems are common in real-world applications across a wide range of domains. Standard classification algorithms are known to perform poorly in these cases, since they focus on overall classification accuracy. In addition, we have seen a significant increase of data in recent years, resulting in many large scale rare class problems. In this paper, we focus on nonlinear kernel based classification methods expressed as a regularized loss minimization problem. We address the challenges associated with both rare class problems and large scale learning, by 1) optimizing area under curve of the receiver of operator characteristic in the training process, instead of classification accuracy and 2) using a rare class kernel representation to achieve an efficient time and space algorithm. We call the algorithm RankRC. We provide justifications for the rare class representation and experimentally illustrate the effectiveness of RankRC in test performance, computational complexity, and model robustness. [ABSTRACT FROM PUBLISHER]

Approximate computing is a promising design paradigm for better performance and power efficiency. In this paper, we propose a power efficient framework for analog approximate computing with the emerging metal-oxide resistive switching random-access memory (RRAM) devices. A programmable RRAM-based approximate computing unit (RRAM-ACU ) is introduced first to accelerate approximated computation, and an approximate computing framework with scalability is then proposed on top of the RRAM-ACU. In order to program the RRAM-ACU efficiently, we also present a detailed configuration flow, which includes a customized approximator training scheme, an approximator-parameter-to-RRAM-state mapping algorithm, and an RRAM state tuning scheme. Finally, the proposed RRAM-based computing framework is modeled at system level. A predictive compact model is developed to estimate the configuration overhead of RRAM-ACU and help explore the application scenarios of RRAM-based analog approximate computing. The simulation results on a set of diverse benchmarks demonstrate that, compared with a x86–64 CPU at 2 GHz, the RRAM-ACU is able to achieve 4.06– 196.41 \times speedup and power efficiency of 24.59–567.98 GFLOPS/W with quality loss of 8.72% on average. And the implementation of hierarchical model and X application demonstrates that the proposed RRAM-based approximate computing framework can achieve >12.8 $\times$ power efficiency than its pure digital implementation counterparts (CPU, graphics processing unit, and field- programmable gate arrays). [ABSTRACT FROM PUBLISHER]

Software process modeling (SPM) is an important area of software engineering because it provides a basis for managing, automating, and supporting software process improvement (SPI). Teaching SPM is a challenging task, mainly because it lays great emphasis on theory and offers few practical exercises. Furthermore, as yet few teaching approaches have aimed at teaching SPM by introducing innovative features, such as games. The use of games has mainly been focused on other areas of software engineering, for example software project management. In an attempt to fill this gap, this paper describes a formal experiment carried out to assess the learning effectiveness of a serious game (DesigMPS), designed to support the teaching of SPM, and to compare game-based learning with a project-based learning method. In the DesigMPS game, the student models a software process from an SPI perspective, based on the Brazilian SPI model (MPS.BR). The results indicate that playing the game can have a positive learning effect and results in a greater degree of learning effectiveness than does the project-based learning instructional method. [ABSTRACT FROM PUBLISHER]

In this paper, an adjoint state-space dynamic neural network method for modeling nonlinear circuits and components is presented. This method is used for modeling the transient behavior of the nonlinear electronic and photonic components. The proposed technique is an extension of the existing state-space dynamic neural network (SSDNN) technique. The new method simultaneously adds the derivative information to the training patterns of nonlinear components, allowing the training to be done with less data without sacrificing model accuracy, and, consequently, makes training faster and more efficient. In addition, this method has been formulated such that it can be suitable for the parallel computation. The use of derivative information and parallelization makes training using the proposed technique much faster than the SSDNN. In addition, the models created using the proposed method are much faster to evaluate compared with the conventional models present in traditional circuit simulation tools. The validity of the proposed technique is demonstrated through the transient modeling of the physics-based CMOS driver, commercial NXP’s 74LVC04A inverting buffer, and nonlinear photonic components. [ABSTRACT FROM PUBLISHER]

As various graphics processing unit architectures are deployed across broad computing spectrum from a hand-held or embedded device to a high-performance computing server, OpenCL becomes the de facto standard programming environment for general-purpose computing on graphics processing units. Unlike its CPU counterpart, OpenCL has several distinct features such as its disciplined memory model, which is partially inherited from conventional 3D graphics programming models. On the other hand, due to ever increasing memory bandwidth pressure and low power requirement, the capacity of on-chip caches in GPUs keeps increasing over time. Given such trends, we believe that we have interesting programming model/architecture co-optimization opportunities, in particular, how to energy-efficiently utilize large on-chip caches for GPUs. In this paper, as a showcase, we study the characteristics of the OpenCL memory model and propose a technique called GPU Region-aware energy-efficient non-inclusive cache hierarchy, or GREEN cache hierarchy. With the GREEN cache, our simulation results show that we can save 56 percent of dynamic energy in the L1 cache, 39 percent of dynamic energy in the L2 cache, and 50 percent of leakage energy in the L2 cache with practically no performance degradation and off-chip access increases. [ABSTRACT FROM PUBLISHER]

Increasingly, aiming to contain their rapidly growing energy expenditures, commercial buildings are equipped to respond to utility’s demand and price signals. Such smart energy consumption, however, heavily relies on accurate short-term energy load forecasting, such as hourly predictions for the next $n$ $(n\ge 2)$ hours. To attain sufficient accuracy for these predictions, it is important to exploit the relationships among the $n$ estimated outputs. This paper treats such multi-steps ahead regression task as a sequence labeling (regression) problem, and adopts the continuous conditional random fields (CCRF) to explicitly model these interconnected outputs. In particular, we improve the CCRF’s computation complexity and predictive accuracy with two novel strategies. First, we employ two tridiagonal matrix computation techniques to significantly speed up the CCRF’s training and inference. These techniques tackle the cubic computational cost required by the matrix inversion calculations in the training and inference of the CCRF, resulting in linear complexity for these matrix operations. Second, we address the CCRF’s weak feature constraint problem with a novel multi-target edge function, thus boosting the CCRF’s predictive performance. The proposed multi-target feature is able to convert the relationship of related outputs with continuous values into a set of “sub-relationships”, each providing more specific feature constraints for the interplays of the related outputs. We applied the proposed approach to two real-world energy load prediction systems: one for electricity demand and another for gas usage. Our experimental results show that the proposed strategy can meaningfully reduce the predictive error for the two systems, in terms of mean absolute percentage error and root mean square error, when compared with three benchmarking methods. Promisingly, the relative error reduction achieved by our CCRF model was up to 50 percent. [ABSTRACT FROM PUBLISHER]

Detailed segmentation of the vertebrae is an important pre-requisite in various applications of image-based spine assessment, surgery and biomechanical modeling. In particular, accurate segmentation of the processes is required for image-guided interventions, for example for optimal placement of bone grafts between the transverse processes. Furthermore, the geometry of the processes is now required in musculoskeletal models due to their interaction with the muscles and ligaments. In this paper, we present a new method for detailed segmentation of both the vertebral bodies and processes based on statistical shape decomposition and conditional models. The proposed technique is specifically developed with the aim to handle the complex geometry of the processes and the large variability between individuals. The key technical novelty in this work is the introduction of a part-based statistical decomposition of the vertebrae, such that the complexity of the subparts is effectively reduced, and model specificity is increased. Subsequently, in order to maintain the statistical and anatomic coherence of the ensemble, conditional models are used to model the statistical inter-relationships between the different subparts. For shape reconstruction and segmentation, a robust model fitting procedure is used to exclude improbable inter-part relationships in the estimation of the shape parameters. Segmentation results based on a dataset of 30 healthy CT scans and a dataset of 10 pathological scans show a point-to-surface error improvement of 20% and 17% respectively, and the potential of the proposed technique for detailed vertebral modeling. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach to enable frequent computational cycles in online dynamic security assessment by using the terms of the transient energy function (TEF) as input features to a machine learning algorithm. The aim is to train a single classifier that is capable of classifying stable and unstable operating points independent of the contingency. The network is trained based on the current system topology and the loading conditions. The potential of the proposed approach is demonstrated with the New England 39-bus test power system model using the support vector machine as the machine learning technique. It is shown that the classifier can be trained using a small set of data when the terms of the TEF are used as input features. The prediction accuracy of the proposed scheme was tested under the balanced and unbalanced faults with the presence of voltage sensitive and dynamic loads for different operating points. [ABSTRACT FROM PUBLISHER]

This paper focuses on the measure of recommendation stability, which reflects the consistency of recommender system predictions. Stability is a desired property of recommendation algorithms and has important implications on users’ trust and acceptance of recommendations. Prior research has reported that some popular recommendation algorithms can suffer from a high degree of instability. In this study, we explore two scalable, general-purpose meta-algorithmic approaches—based on bagging and iterative smoothing—that can be used in conjunction with different traditional recommendation algorithms to improve their stability. Our experimental results on real-world rating data demonstrate that both approaches can achieve substantially higher stability as compared to the original recommendation algorithms. Furthermore, perhaps as importantly, the proposed approaches not only do not sacrifice the predictive accuracy in order to improve recommendation stability, but are actually able to provide additional accuracy improvements. [ABSTRACT FROM AUTHOR]

Pose estimation is a task with wide application prospects in computer vision, which remains a challenging problem. In this paper, a novel pose estimation algorithm is proposed on the basis of pose clustering and body-part candidates recombination. Different from most previous methods with a single pictorial structure (PS) model, we generate mixture PS models based on clusters of the poses to achieve more faithful appearances and spatial relations estimation within each cluster. In addition, to address the problems of individual body-part false detection and double-counting, we extract some of the best estimation results in the optimal clustered model as the candidates of body parts and recombine them by solving a constrained maximization problem. Experiments on a public challenging data set show that our method is more accurate than the state-of-the-art algorithms and performs effectively in tackling the double-counting phenomena. [ABSTRACT FROM PUBLISHER]

Defect detection and classification in semiconductor wafers has received an increasing attention from both industry and academia alike. Wafer defects are a serious problem that could cause massive losses to the companies’ yield. The defects occur as a result of a lengthy and complex fabrication process involving hundreds of stages, and they can create unique patterns. If these patterns were to be identified and classified correctly, then the root of the fabrication problem can be recognized and eventually resolved. Machine learning (ML) techniques have been widely accepted and are well suited for such classification-/identification problems. However, none of the existing ML model’s performance exceeds 96% in identification accuracy for such tasks. In this paper, we develop a state-of-the-art classifying algorithm using multiple ML techniques, relying on a general-regression-network-based consensus learning model along with a powerful randomization technique. We compare our proposed method with the widely used ML models in terms of model accuracy, stability, and time complexity. Our method has proved to be more accurate and stable as compared to any of the existing algorithms reported in the literature, achieving its accuracy of 99.8%, stability of 1.128, and TBM of 15.8 s. [ABSTRACT FROM AUTHOR]

Recently, most background modeling approaches represent distributions of background changes by using parametric models such as Gaussian mixture models. Because of significant illumination changes and dynamic moving backgrounds with time, variations of background changes are hard to be modeled by parametric background models. Moreover, how to efficiently and effectively update parameters of parametric models to reflect background changes remains a problem. In this paper, we propose a novel coarse-to-fine detection theory algorithm to extract foreground objects on the basis of nonparametric background and foreground models represented by binary descriptors. We update background and foreground models by a first-in–first-out strategy to maintain the most recent observed background and foreground instances. As shown in the experiments, our method can achieve better foreground extraction results and fewer false alarms of surveillance videos with lighting changes and dynamic backgrounds in both collected and CDnet 2012 benchmark data sets. [ABSTRACT FROM AUTHOR]

Real-world tasks in computer vision often touch upon open set recognition: multi-class recognition with incomplete knowledge of the world and many unknown inputs. Recent work on this problem has proposed a model incorporating an open space risk term to account for the space beyond the reasonable support of known classes. This paper extends the general idea of open space risk limiting classification to accommodate non-linear classifiers in a multi-class setting. We introduce a new open set recognition model called compact abating probability (CAP), where the probability of class membership decreases in value (abates) as points move from known data toward open space. We show that CAP models improve open set recognition for multiple algorithms. Leveraging the CAP formulation, we go on to describe the novel Weibull-calibrated SVM (W-SVM) algorithm, which combines the useful properties of statistical extreme value theory for score calibration with one-class and binary support vector machines. Our experiments show that the W-SVM is significantly better for open set object detection and OCR problems when compared to the state-of-the-art for the same tasks. [ABSTRACT FROM AUTHOR]

Lung field segmentation in the posterior–anterior (PA) chest radiograph is important for pulmonary disease diagnosis and hemodialysis treatment. Due to high shape variation and boundary ambiguity, accurate lung field segmentation from chest radiograph is still a challenging task. To tackle these challenges, we propose a joint shape and appearance sparse learning method for robust and accurate lung field segmentation. The main contributions of this paper are: 1) a robust shape initialization method is designed to achieve an initial shape that is close to the lung boundary under segmentation; 2) a set of local sparse shape composition models are built based on local lung shape segments to overcome the high shape variations; 3) a set of local appearance models are similarly adopted by using sparse representation to capture the appearance characteristics in local lung boundary segments, thus effectively dealing with the lung boundary ambiguity; 4) a hierarchical deformable segmentation framework is proposed to integrate the scale-dependent shape and appearance information together for robust and accurate segmentation. Our method is evaluated on 247 PA chest radiographs in a public dataset. The experimental results show that the proposed local shape and appearance models outperform the conventional shape and appearance models. Compared with most of the state-of-the-art lung field segmentation methods under comparison, our method also shows a higher accuracy, which is comparable to the inter-observer annotation variation. [ABSTRACT FROM AUTHOR]

We study the problem of online multitask learning for solving multiple related classification tasks in parallel, aiming at classifying every sequence of data received by each task accurately and efficiently. One practical example of online multitask learning is the micro-blog sentiment detection on a group of users, which classifies micro-blog posts generated by each user into emotional or non-emotional categories. This particular online learning task is challenging for a number of reasons. First of all, to meet the critical requirements of online applications, a highly efficient and scalable classification solution that can make immediate predictions with low learning cost is needed. This requirement leaves conventional batch learning algorithms out of consideration. Second, classical classification methods, be it batch or online, often encounter a dilemma when applied to a group of tasks, i.e., on one hand, a single classification model trained on the entire collection of data from all tasks may fail to capture characteristics of individual task; on the other hand, a model trained independently on individual tasks may suffer from insufficient training data. To overcome these challenges, in this paper, we propose a collaborative online multitask learning method, which learns a global model over the entire data of all tasks. At the same time, individual models for multiple related tasks are jointly inferred by leveraging the global model through a collaborative online learning approach. We illustrate the efficacy of the proposed technique on a synthetic dataset. We also evaluate it on three real-life problems—spam email filtering, bioinformatics data classification, and micro-blog sentiment detection. Experimental results show that our method is effective and scalable at the online classification of multiple related tasks. [ABSTRACT FROM PUBLISHER]