arXiv subjects

Xiang Chen

Publications and source records attributed to Xiang Chen.

At least 523 records · Page 29Linked to original sources

arXiv2019

Tiny but Accurate: A Pruned, Quantized and Optimized Memristor Crossbar Framework for Ultra Efficient DNN Implementation

The state-of-art DNN structures involve intensive computation and high memory storage. To mitigate the challenges, the memristor crossbar array has emerged as an intrinsically suitable matrix computation and low-power acceleration framework for DNN applications. However, the high accuracy solution for extreme model compression on memristor crossbar array architecture is still waiting for unraveling. In this paper, we propose a memristor-based DNN framework which combines both structured weight pruning and quantization by incorporating alternating direction method of multipliers (ADMM) algorithm for better pruning and quantization performance. We also discover the non-optimality of the ADMM solution in weight pruning and the unused data path in a structured pruned model. Motivated by these discoveries, we design a software-hardware co-optimization framework which contains the first proposed Network Purification and Unused Path Removal algorithms targeting on post-processing a structured pruned model after ADMM steps. By taking memristor hardware constraints into our whole framework, we achieve extreme high compression ratio on the state-of-art neural network structures with minimum accuracy loss. For quantizing structured pruned model, our framework achieves nearly no accuracy loss after quantizing weights to 8-bit memristor weight representation. We share our models at anonymous link https://bit.ly/2VnMUy0.

eess.SP↗

arXiv2019

Multi-stage Deep Classifier Cascades for Open World Recognition

At present, object recognition studies are mostly conducted in a closed lab setting with classes in test phase typically in training phase. However, real-world problem is far more challenging because: i) new classes unseen in the training phase can appear when predicting; ii) discriminative features need to evolve when new classes emerge in real time; and iii) instances in new classes may not follow the "independent and identically distributed" (iid) assumption. Most existing work only aims to detect the unknown classes and is incapable of continuing to learn newer classes. Although a few methods consider both detecting and including new classes, all are based on the predefined handcrafted features that cannot evolve and are out-of-date for characterizing emerging classes. Thus, to address the above challenges, we propose a novel generic end-to-end framework consisting of a dynamic cascade of classifiers that incrementally learn their dynamic and inherent features. The proposed method injects dynamic elements into the system by detecting instances from unknown classes, while at the same time incrementally updating the model to include the new classes. The resulting cascade tree grows by adding a new leaf node classifier once a new class is detected, and the discriminative features are updated via an end-to-end learning strategy. Experiments on two real-world datasets demonstrate that our proposed method outperforms existing state-of-the-art methods.

cs.LG↗

arXiv2019

Revisiting Heterogeneous Defect Prediction: How Far Are We?

Until now, researchers have proposed several novel heterogeneous defect prediction HDP methods with promising performance. To the best of our knowledge, whether HDP methods can perform significantly better than unsupervised methods has not yet been thoroughly investigated. In this article, we perform a replication study to have a holistic look in this issue. In particular, we compare state-of-the-art five HDP methods with five unsupervised methods. Final results surprisingly show that these HDP methods do not perform significantly better than some of unsupervised methods (especially the simple unsupervised methods proposed by Zhou et al.) in terms of two non-effort-aware performance measures and four effort-aware performance measures. Then, we perform diversity analysis on defective modules via McNemar's test and find the prediction diversity is more obvious when the comparison is performed between the HDP methods and the unsupervised methods than the comparisons only between the HDP methods or between the unsupervised methods. This shows the HDP methods and the unsupervised methods are complementary to each other in identifying defective models to some extent. Finally, we investigate the feasibility of five HDP methods by considering two satisfactory criteria recommended by previous CPDP studies and find the satisfactory ratio of these HDP methods is still pessimistic. The above empirical results implicate there is still a long way for heterogeneous defect prediction to go. More effective HDP methods need to be designed and the unsupervised methods should be considered as baselines.

cs.SE↗

arXiv2019

Interpreting and Evaluating Neural Network Robustness

Recently, adversarial deception becomes one of the most considerable threats to deep neural networks. However, compared to extensive research in new designs of various adversarial attacks and defenses, the neural networks' intrinsic robustness property is still lack of thorough investigation. This work aims to qualitatively interpret the adversarial attack and defense mechanism through loss visualization, and establish a quantitative metric to evaluate the neural network model's intrinsic robustness. The proposed robustness metric identifies the upper bound of a model's prediction divergence in the given domain and thus indicates whether the model can maintain a stable prediction. With extensive experiments, our metric demonstrates several advantages over conventional adversarial testing accuracy based robustness estimation: (1) it provides a uniformed evaluation to models with different structures and parameter scales; (2) it over-performs conventional accuracy based robustness estimation and provides a more reliable evaluation that is invariant to different test settings; (3) it can be fast generated without considerable testing cost.

cs.LG↗

arXiv2019

FTGAN: A Fully-trained Generative Adversarial Networks for Text to Face Generation

As a sub-domain of text-to-image synthesis, text-to-face generation has huge potentials in public safety domain. With lack of dataset, there are almost no related research focusing on text-to-face synthesis. In this paper, we propose a fully-trained Generative Adversarial Network (FTGAN) that trains the text encoder and image decoder at the same time for fine-grained text-to-face generation. With a novel fully-trained generative network, FTGAN can synthesize higher-quality images and urge the outputs of the FTGAN are more relevant to the input sentences. In addition, we build a dataset called SCU-Text2face for text-to-face synthesis. Through extensive experiments, the FTGAN shows its superiority in boosting both generated images' quality and similarity to the input descriptions. The proposed FTGAN outperforms the previous state of the art, boosting the best reported Inception Score to 4.63 on the CUB dataset. On SCU-text2face, the face images generated by our proposed FTGAN just based on the input descriptions is of average 59% similarity to the ground-truth, which set a baseline for text-to-face synthesis.

cs.CV↗

arXiv2019

Symmetries and Local Conservation Laws of Variational Schemes for the Surface Plasmon Polaritons

The relation between symmetries and local conservation laws, known as Noether's theorem, plays an important role in modern theoretical physics. As a discrete analog of the differentiable physical system, a good numerical scheme should admit the discrete local conservation laws and inherent mathematical structures. A class of variational schemes constructed for the hydrodynamic-electrodynamic model of lossless free-electron gas in a quasi-neutral background shows good properties in secular simulations of surface plasmon polaritons [Q. Chen et al., Phys. Rev. E 99, 023313 (2019)]. We show the discrete local conservation laws admitted by these schemes. Based on the gauge symmetry of the discrete action functional, a discrete charge conservation law is realized locally, which is consistent with the discrete Euler-Lagrange equations obtained from the variational schemes. Based on the discrete Euler-Lagrange equations, discrete local momentum and energy conservation laws are derived directly, which are rigorous in theory. The preservation of the discrete local conservation laws and Lagrangian symplectic structure ensure that the numerical scheme is correct in physics.

physics.comp-ph↗

arXiv2019

DeepCount: Crowd Counting with WiFi via Deep Learning

Recently, the research of wireless sensing has achieved more intelligent results, and the intelligent sensing of human location and activity can be realized by means of WiFi devices. However, most of the current human environment perception work is limited to a single person's environment, because the environment in which multiple people exist is more complicated than the environment in which a single person exists. In order to solve the problem of human behavior perception in a multi-human environment, we first proposed a solution to achieve crowd counting (inferred population) using deep learning in a closed environment with WIFI signals - DeepCout, which is the first in a multi-human environment. step. Since the use of WiFi to directly count the crowd is too complicated, we use deep learning to solve this problem, use Convolutional Neural Network(CNN) to automatically extract the relationship between the number of people and the channel, and use Long Short Term Memory(LSTM) to resolve the dependencies of number of people and Channel State Information(CSI) . To overcome the massive labelled data required by deep learning method, we add an online learning mechanism to determine whether or not someone is entering/leaving the room by activity recognition model, so as to correct the deep learning model in the fine-tune stage, which, in turn, reduces the required training data and make our method evolving over time. The system of DeepCount is performed and evaluated on the commercial WiFi devices. By massive training samples, our end-to-end learning approach can achieve an average of 86.4% prediction accuracy in an environment of up to 5 people. Meanwhile, by the amendment mechanism of the activity recognition model to judge door switch to get the variance of crowd to amend deep learning predicted results, the accuracy is up to 90%.

cs.LG↗

arXiv2019

Variational Schemes and Geometric Simulations for a Hydrodynamic-Electrodynamic Model of Surface Plasmon Polaritons

A class of variational schemes for the hydrodynamic-electrodynamic model of lossless free-electron gas in a quasineutral background is developed for high-quality simulations of surface plasmon polaritons. The Lagrangian density of lossless free-electron gas with a self-consistent electromagnetic field is established, and the dynamical equations with the associated constraints are obtained via a variational principle. Based on discrete exterior calculus, the action functional of this system is discretized and minimized to obtain the discrete dynamics. Newton-Raphson iteration and the biconjugate gradient stabilized method are equipped as a hybrid nonlinear-linear algebraic solver. Instead of discretizing the partial differential equations, the variational schemes have better numerical properties in secular simulations, as they preserve the discrete Lagrangian symplectic structure, gauge symmetry, and general energy-momentum density. Two numerical experiments were performed. The numerical results reproduce characteristic dispersion relations of bulk plasmons and surface plasmon polaritons, and the numerical errors of conserved quantities in all experiments are bounded by a small value after long-term simulations.

physics.comp-ph↗

arXiv2019

Reducing Age-of-Information for Computation-Intensive Messages via Packet Replacement

Freshness of data is an important performance metric for real-time applications, which can be measured by age-of-information. For computation-intensive messages, the embedded information is not available until being computed. In this paper, we study the age-of-information for computation-intensive messages, which are firstly transmitted to a mobile edge server, and then processed in the edge server to extract the embedded information. The packet generation follows zero-wait policy, by which a new packet is generated when the last one is just delivered to the edge server. The queue in front of the edge server adopts one-packet-buffer replacement policy, meaning that only the latest received packet is preserved. We derive the expression of average age-of-information for exponentially distributed transmission time and computing time. With packet replacement, the average age is reduced compared with the case without packet replacement, especially when the transmission rate is close to or greater than the computing rate.

cs.IT↗

arXiv2019

Age-of-Information for Computation-Intensive Messages in Mobile Edge Computing

Age-of-information (AoI) is a novel metric that measures the freshness of information in status update scenarios. It is essential for real-time applications to transmit status update packets to the destination node as timely as possible. However, for some applications, status information embedded in the packets is not revealed until complicated data processing, which is computational expensive and time consuming. As mobile edge server has sufficient computational resource and is placed close to users, mobile edge computing (MEC) is expected to reduce age for computation-intensive messages. In this paper, we study the AoI for computation-intensive data in MEC, and consider two schemes: local computing by user itself and remote computing at MEC server. The two computing models are unified into a two-node tandem queuing model. Zero-wait policy is adopted, i.e., a new message is generated once the previous one leaves the first node. We consider exponentially distributed service time and infinite queue size, and hence, the second node can be seen as a First-Come-First-Served (FCFS) M/M/1 system. Closed-form average AoI is derived for the two computing schemes. The region where remote computing outperforms local computing is characterized. Simulation results show that the remote computing is greatly superior to the local computing when the remote computing rate is large enough, and that there exists an optimal transmission rate so that remote computing is better than local computing for a largest range.

cs.DC↗

arXiv2018

Background Subtraction with Real-time Semantic Segmentation

Accurate and fast foreground object extraction is very important for object tracking and recognition in video surveillance. Although many background subtraction (BGS) methods have been proposed in the recent past, it is still regarded as a tough problem due to the variety of challenging situations that occur in real-world scenarios. In this paper, we explore this problem from a new perspective and propose a novel background subtraction framework with real-time semantic segmentation (RTSS). Our proposed framework consists of two components, a traditional BGS segmenter $\mathcal{B}$ and a real-time semantic segmenter $\mathcal{S}$. The BGS segmenter $\mathcal{B}$ aims to construct background models and segments foreground objects. The real-time semantic segmenter $\mathcal{S}$ is used to refine the foreground segmentation outputs as feedbacks for improving the model updating accuracy. $\mathcal{B}$ and $\mathcal{S}$ work in parallel on two threads. For each input frame $I_t$, the BGS segmenter $\mathcal{B}$ computes a preliminary foreground/background (FG/BG) mask $B_t$. At the same time, the real-time semantic segmenter $\mathcal{S}$ extracts the object-level semantics ${S}_t$. Then, some specific rules are applied on ${B}_t$ and ${S}_t$ to generate the final detection ${D}_t$. Finally, the refined FG/BG mask ${D}_t$ is fed back to update the background model. Comprehensive experiments evaluated on the CDnet 2014 dataset demonstrate that our proposed method achieves state-of-the-art performance among all unsupervised background subtraction methods while operating at real-time, and even performs better than some deep learning based supervised algorithms. In addition, our proposed framework is very flexible and has the potential for generalization.

cs.CV↗

arXiv2018

Distilling Critical Paths in Convolutional Neural Networks

Neural network compression and acceleration are widely demanded currently due to the resource constraints on most deployment targets. In this paper, through analyzing the filter activation, gradients, and visualizing the filters' functionality in convolutional neural networks, we show that the filters in higher layers learn extremely task-specific features, which are exclusive for only a small subset of the overall tasks, or even a single class. Based on such findings, we reveal the critical paths of information flow for different classes. And by their intrinsic property of exclusiveness, we propose a critical path distillation method, which can effectively customize the convolutional neural networks to small ones with much smaller model size and less computation.

cs.CV↗

arXiv2018

Progressive Weight Pruning of Deep Neural Networks using ADMM

Deep neural networks (DNNs) although achieving human-level performance in many domains, have very large model size that hinders their broader applications on edge computing devices. Extensive research work have been conducted on DNN model compression or pruning. However, most of the previous work took heuristic approaches. This work proposes a progressive weight pruning approach based on ADMM (Alternating Direction Method of Multipliers), a powerful technique to deal with non-convex optimization problems with potentially combinatorial constraints. Motivated by dynamic programming, the proposed method reaches extremely high pruning rate by using partial prunings with moderate pruning rates. Therefore, it resolves the accuracy degradation and long convergence time problems when pursuing extremely high pruning ratios. It achieves up to 34 times pruning rate for ImageNet dataset and 167 times pruning rate for MNIST dataset, significantly higher than those reached by the literature work. Under the same number of epochs, the proposed method also achieves faster convergence and higher compression rates. The codes and pruned DNN models are released in the link bit.ly/2zxdlss

cs.LG↗

arXiv2018

Linear-Quadratic Mean Field Control: The Hamiltonian Matrix and Invariant Subspace Method

This paper studies the existence and uniqueness of a solution to linear quadratic (LQ) mean field social optimization problems with uniform agents. We exploit a Hamiltonian matrix structure of the associated ordinary differential equation (ODE) system and apply a subspace decomposition method to find the solution. This approach is effective for both the existence analysis and numerical computations. We further extend the decomposition method to LQ mean field games.

math.OC↗

arXiv2018

Demystifying Neural Network Filter Pruning

Based on filter magnitude ranking (e.g. L1 norm), conventional filter pruning methods for Convolutional Neural Networks (CNNs) have been proved with great effectiveness in computation load reduction. Although effective, these methods are rarely analyzed in a perspective of filter functionality. In this work, we explore the filter pruning and the retraining through qualitative filter functionality interpretation. We find that the filter magnitude based method fails to eliminate the filters with repetitive functionality. And the retraining phase is actually used to reconstruct the remained filters for functionality compensation for the wrongly-pruned critical filters. With a proposed functionality-oriented pruning method, we further testify that, by precisely addressing the filter functionality redundancy, a CNN can be pruned without considerable accuracy drop, and the retraining phase is unnecessary.

cs.CV↗

arXiv2018

Interpreting Adversarial Robustness: A View from Decision Surface in Input Space

One popular hypothesis of neural network generalization is that the flat local minima of loss surface in parameter space leads to good generalization. However, we demonstrate that loss surface in parameter space has no obvious relationship with generalization, especially under adversarial settings. Through visualizing decision surfaces in both parameter space and input space, we instead show that the geometry property of decision surface in input space correlates well with the adversarial robustness. We then propose an adversarial robustness indicator, which can evaluate a neural network's intrinsic robustness property without testing its accuracy under adversarial attacks. Guided by it, we further propose our robust training method. Without involving adversarial training, our method could enhance network's intrinsic adversarial robustness against various adversarial attacks.

cs.LG↗

arXiv2018

A Hybrid Neural Network Framework and Application to Radar Automatic Target Recognition

Deep neural networks (DNNs) have found applications in diverse signal processing (SP) problems. Most efforts either directly adopt the DNN as a black-box approach to perform certain SP tasks without taking into account of any known properties of the signal models, or insert a pre-defined SP operator into a DNN as an add-on data processing stage. This paper presents a novel hybrid-NN framework in which one or more SP layers are inserted into the DNN architecture in a coherent manner to enhance the network capability and efficiency in feature extraction. These SP layers are properly designed to make good use of the available models and properties of the data. The network training algorithm of hybrid-NN is designed to actively involve the SP layers in the learning goal, by simultaneously optimizing both the weights of the DNN and the unknown tuning parameters of the SP operators. The proposed hybrid-NN is tested on a radar automatic target recognition (ATR) problem. It achieves high validation accuracy of 96\% with 5,000 training images in radar ATR. Compared with ordinary DNN, hybrid-NN can markedly reduce the required amount of training data and improve the learning performance.

eess.SP↗

arXiv2018

HASP: A High-Performance Adaptive Mobile Security Enhancement Against Malicious Speech Recognition

Nowadays, machine learning based Automatic Speech Recognition (ASR) technique has widely spread in smartphones, home devices, and public facilities. As convenient as this technology can be, a considerable security issue also raises -- the users' speech content might be exposed to malicious ASR monitoring and cause severe privacy leakage. In this work, we propose HASP -- a high-performance security enhancement approach to solve this security issue on mobile devices. Leveraging ASR systems' vulnerability to the adversarial examples, HASP is designed to cast human imperceptible adversarial noises to real-time speech and effectively perturb malicious ASR monitoring by increasing the Word Error Rate (WER). To enhance the practical performance on mobile devices, HASP is also optimized for effective adaptation to the human speech characteristics, environmental noises, and mobile computation scenarios. The experiments show that HASP can achieve optimal real-time security enhancement: it can lead an average WER of 84.55% for perturbing the malicious ASR monitoring, and the data processing speed is 15x to 40x faster compared to the state-of-the-art methods. Moreover, HASP can effectively perturb various ASR systems, demonstrating a strong transferability.

cs.CR↗