arXiv subjects

Tong Zhang

Publications and source records attributed to Tong Zhang.

At least 631 records · Page 35Linked to original sources

arXiv2018

When Work Matters: Transforming Classical Network Structures to Graph CNN

Numerous pattern recognition applications can be formed as learning from graph-structured data, including social network, protein-interaction network, the world wide web data, knowledge graph, etc. While convolutional neural network (CNN) facilitates great advances in gridded image/video understanding tasks, very limited attention has been devoted to transform these successful network structures (including Inception net, Residual net, Dense net, etc.) to establish convolutional networks on graph, due to its irregularity and complexity geometric topologies (unordered vertices, unfixed number of adjacent edges/vertices). In this paper, we aim to give a comprehensive analysis of when work matters by transforming different classical network structures to graph CNN, particularly in the basic graph recognition problem. Specifically, we firstly review the general graph CNN methods, especially in its spectral filtering operation on the irregular graph data. We then introduce the basic structures of ResNet, Inception and DenseNet into graph CNN and construct these network structures on graph, named as G_ResNet, G_Inception, G_DenseNet. In particular, it seeks to help graph CNNs by shedding light on how these classical network structures work and providing guidelines for choosing appropriate graph network frameworks. Finally, we comprehensively evaluate the performance of these different network structures on several public graph datasets (including social networks and bioinformatic datasets), and demonstrate how different network structures work on graph CNN in the graph recognition task.

cs.LG↗

arXiv2018

Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization

Large-scale distributed optimization is of great importance in various applications. For data-parallel based distributed learning, the inter-node gradient communication often becomes the performance bottleneck. In this paper, we propose the error compensated quantized stochastic gradient descent algorithm to improve the training efficiency. Local gradients are quantized to reduce the communication overhead, and accumulated quantization error is utilized to speed up the convergence. Furthermore, we present theoretical analysis on the convergence behaviour, and demonstrate its advantage over competitors. Extensive experiments indicate that our algorithm can compress gradients by a factor of up to two magnitudes without performance degradation.

cs.CV↗

arXiv2018

Composite Functional Gradient Learning of Generative Adversarial Models

This paper first presents a theory for generative adversarial methods that does not rely on the traditional minimax formulation. It shows that with a strong discriminator, a good generator can be learned so that the KL divergence between the distributions of real data and generated data improves after each functional gradient step until it converges to zero. Based on the theory, we propose a new stable generative adversarial method. A theoretical insight into the original GAN from this new viewpoint is also provided. The experiments on image generation show the effectiveness of our new method.

stat.ML↗

arXiv2018

An Algorithmic Framework of Variable Metric Over-Relaxed Hybrid Proximal Extra-Gradient Method

We propose a novel algorithmic framework of Variable Metric Over-Relaxed Hybrid Proximal Extra-gradient (VMOR-HPE) method with a global convergence guarantee for the maximal monotone operator inclusion problem. Its iteration complexities and local linear convergence rate are provided, which theoretically demonstrate that a large over-relaxed step-size contributes to accelerating the proposed VMOR-HPE as a byproduct. Specifically, we find that a large class of primal and primal-dual operator splitting algorithms are all special cases of VMOR-HPE. Hence, the proposed framework offers a new insight into these operator splitting algorithms. In addition, we apply VMOR-HPE to the Karush-Kuhn-Tucker (KKT) generalized equation of linear equality constrained multi-block composite convex optimization, yielding a new algorithm, namely nonsymmetric Proximal Alternating Direction Method of Multipliers with a preconditioned Extra-gradient step in which the preconditioned metric is generated by a blockwise Barzilai-Borwein line search technique (PADMM-EBB). We also establish iteration complexities of PADMM-EBB in terms of the KKT residual. Finally, we apply PADMM-EBB to handle the nonnegative dual graph regularized low-rank representation problem. Promising results on synthetic and real datasets corroborate the efficacy of PADMM-EBB.

math.OC↗

arXiv2018

Safe Element Screening for Submodular Function Minimization

Submodular functions are discrete analogs of convex functions, which have applications in various fields, including machine learning and computer vision. However, in large-scale applications, solving Submodular Function Minimization (SFM) problems remains challenging. In this paper, we make the first attempt to extend the emerging technique named screening in large-scale sparse learning to SFM for accelerating its optimization process. We first conduct a careful studying of the relationships between SFM and the corresponding convex proximal problems, as well as the accurate primal optimum estimation of the proximal problems. Relying on this study, we subsequently propose a novel safe screening method to quickly identify the elements guaranteed to be included (we refer to them as active) or excluded (inactive) in the final optimal solution of SFM during the optimization process. By removing the inactive elements and fixing the active ones, the problem size can be dramatically reduced, leading to great savings in the computational cost without sacrificing any accuracy. To the best of our knowledge, the proposed method is the first screening method in the fields of SFM and even combinatorial optimization, thus pointing out a new direction for accelerating SFM algorithms. Experiment results on both synthetic and real datasets demonstrate the significant speedups gained by our approach.

stat.ML↗

arXiv2018

End-to-end Active Object Tracking via Reinforcement Learning

We study active object tracking, where a tracker takes as input the visual observation (i.e., frame sequence) and produces the camera control signal (e.g., move forward, turn left, etc.). Conventional methods tackle the tracking and the camera control separately, which is challenging to tune jointly. It also incurs many human efforts for labeling and many expensive trial-and-errors in realworld. To address these issues, we propose, in this paper, an end-to-end solution via deep reinforcement learning, where a ConvNet-LSTM function approximator is adopted for the direct frame-toaction prediction. We further propose an environment augmentation technique and a customized reward function, which are crucial for a successful training. The tracker trained in simulators (ViZDoom, Unreal Engine) shows good generalization in the case of unseen object moving path, unseen object appearance, unseen background, and distracting object. It can restore tracking when occasionally losing the target. With the experiments over the VOT dataset, we also find that the tracking ability, obtained solely from simulators, can potentially transfer to real-world scenarios.

cs.CV↗

arXiv2018

Fine-grained Video Attractiveness Prediction Using Multimodal Deep Learning on a Large Real-world Dataset

Nowadays, billions of videos are online ready to be viewed and shared. Among an enormous volume of videos, some popular ones are widely viewed by online users while the majority attract little attention. Furthermore, within each video, different segments may attract significantly different numbers of views. This phenomenon leads to a challenging yet important problem, namely fine-grained video attractiveness prediction. However, one major obstacle for such a challenging problem is that no suitable benchmark dataset currently exists. To this end, we construct the first fine-grained video attractiveness dataset, which is collected from one of the most popular video websites in the world. In total, the constructed FVAD consists of 1,019 drama episodes with 780.6 hours covering different categories and a wide variety of video contents. Apart from the large amount of videos, hundreds of millions of user behaviors during watching videos are also included, such as "view counts", "fast-forward", "fast-rewind", and so on, where "view counts" reflects the video attractiveness while other engagements capture the interactions between the viewers and videos. First, we demonstrate that video attractiveness and different engagements present different relationships. Second, FVAD provides us an opportunity to study the fine-grained video attractiveness prediction problem. We design different sequential models to perform video attractiveness prediction by relying solely on video contents. The sequential models exploit the multimodal relationships between visual and audio components of the video contents at different levels. Experimental results demonstrate the effectiveness of our proposed sequential models with different visual and audio representations, the necessity of incorporating the two modalities, and the complementary behaviors of the sequential prediction models at different levels.

cs.CV↗

arXiv2018

Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective

The success of current deep saliency detection methods heavily depends on the availability of large-scale supervision in the form of per-pixel labeling. Such supervision, while labor-intensive and not always possible, tends to hinder the generalization ability of the learned models. By contrast, traditional handcrafted features based unsupervised saliency detection methods, even though have been surpassed by the deep supervised methods, are generally dataset-independent and could be applied in the wild. This raises a natural question that "Is it possible to learn saliency maps without using labeled data while improving the generalization ability?". To this end, we present a novel perspective to unsupervised saliency detection through learning from multiple noisy labeling generated by "weak" and "noisy" unsupervised handcrafted saliency methods. Our end-to-end deep learning framework for unsupervised saliency detection consists of a latent saliency prediction module and a noise modeling module that work collaboratively and are optimized jointly. Explicit noise modeling enables us to deal with noisy saliency maps in a probabilistic way. Extensive experimental results on various benchmarking datasets show that our model not only outperforms all the unsupervised saliency methods with a large margin but also achieves comparable performance with the recent state-of-the-art supervised deep saliency methods.

cs.CV↗

arXiv2018

Tensor graph convolutional neural network

In this paper, we propose a novel tensor graph convolutional neural network (TGCNN) to conduct convolution on factorizable graphs, for which here two types of problems are focused, one is sequential dynamic graphs and the other is cross-attribute graphs. Especially, we propose a graph preserving layer to memorize salient nodes of those factorized subgraphs, i.e. cross graph convolution and graph pooling. For cross graph convolution, a parameterized Kronecker sum operation is proposed to generate a conjunctive adjacency matrix characterizing the relationship between every pair of nodes across two subgraphs. Taking this operation, then general graph convolution may be efficiently performed followed by the composition of small matrices, which thus reduces high memory and computational burden. Encapsuling sequence graphs into a recursive learning, the dynamics of graphs can be efficiently encoded as well as the spatial layout of graphs. To validate the proposed TGCNN, experiments are conducted on skeleton action datasets as well as matrix completion dataset. The experiment results demonstrate that our method can achieve more competitive performance with the state-of-the-art methods.

cs.CV↗

arXiv2018

An Analysis on Caching Placement for Millimeter/Micro Wave Hybrid Networks

In this paper, we consider a hybrid millimeter wave (mmWave) and micro wave ($μ$Wave) network from the perspective of \emph{wireless caching} and study the optimal probabilistic content/file caching placement at desirable base stations (BSs) using a stochastic geometric framework. Considering the average success probability (ASP) of file delivery as the performance metric, we derive expressions for the association probability of the typical user to the mmWave and $μ$Wave networks. Accordingly, we provide an upper bound for the ASP of file delivery and formulate the content caching placement scheme as an optimization problem with respect to caching probabilities, that jointly optimizes the ASP of file delivery considering both content placement and delivery phases. In particular, we consider the caching placement strategy under both noise-limited and interference-limited environments. We numerically evaluate the performance of the proposed caching schemes under essential factors, such as blockages in the mmWave network, cluster radius, BS density, and path loss and compare it with uniform caching placement, caching $M$ most popular contents, and random caching placement. Numerical results demonstrate the superiority of the proposed caching scheme over others, albeit certain trade-offs.

eess.SP↗

arXiv2018

Relative Clifford inequality for varieties fibered by curves

We prove a sharp relative Clifford inequality for relatively special divisors on varieties fibered by curves. It generalizes the classical Clifford inequality about a single curve to a family of curves. It yields a geographical inequality for varieties of general type and Albanese-fibered by curves, extending the work of Horikawa, Persson, and Xiao in dimension two to arbitrary dimensions. We also apply it to deduce a slope inequality for some arbitrary dimensional families of curves. It sheds light on the existence of a most general Cornalba-Harris-Xiao type inequality for families of curves. The whole proof is built on a new tree-like filtration for nef divisors on varieties fibered by curves. One key ingredient of the proof is to estimate the sum of all admissible products of nef thresholds with respect to this filtration.

math.AG↗

arXiv2018

Bayesian Model Averaging with Exponentiated Least Square Loss

The model averaging problem is to average multiple models to achieve a prediction accuracy not much worse than that of the best single model in terms of mean squared error. It is known that if the models are misspecified, model averaging is superior to model selection. Specifically, let $n$ be the sample size, then the worst case regret of the former decays at a rate of $O(1/n)$ while the worst case regret of the latter decays at a rate of $O(1/\sqrt{n})$. The recently proposed $Q$-aggregation algorithm \citep{DaiRigZhang12} solves the model averaging problem with the optimal regret of $O(1/n)$ both in expectation and in deviation; however it suffers from two limitations: (1) for continuous dictionary, the proposed greedy algorithm for solving $Q$-aggregation is not applicable; (2) the formulation of $Q$-aggregation appears ad hoc without clear intuition. This paper examines a different approach to model averaging by considering a Bayes estimator for deviation optimal model averaging by using exponentiated least squares loss. We establish a primal-dual relationship of this estimator and that of $Q$-aggregation and propose new greedy procedures that satisfactorily resolve the above mentioned limitations of $Q$-aggregation.

math.ST↗

arXiv2018

Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents

We consider the problem of \emph{fully decentralized} multi-agent reinforcement learning (MARL), where the agents are located at the nodes of a time-varying communication network. Specifically, we assume that the reward functions of the agents might correspond to different tasks, and are only known to the corresponding agent. Moreover, each agent makes individual decisions based on both the information observed locally and the messages received from its neighbors over the network. Within this setting, the collective goal of the agents is to maximize the globally averaged return over the network through exchanging information with their neighbors. To this end, we propose two decentralized actor-critic algorithms with function approximation, which are applicable to large-scale MARL problems where both the number of states and the number of agents are massively large. Under the decentralized structure, the actor step is performed individually by each agent with no need to infer the policies of others. For the critic step, we propose a consensus update via communication over the network. Our algorithms are fully incremental and can be implemented in an online fashion. Convergence analyses of the algorithms are provided when the value functions are approximated within the class of linear functions. Extensive simulation results with both linear and nonlinear function approximations are presented to validate the proposed algorithms. Our work appears to be the first study of fully decentralized MARL algorithms for networked agents with function approximation, with provable convergence guarantees.

cs.LG↗

arXiv2018

On Quadratic Convergence of DC Proximal Newton Algorithm for Nonconvex Sparse Learning in High Dimensions

We propose a DC proximal Newton algorithm for solving nonconvex regularized sparse learning problems in high dimensions. Our proposed algorithm integrates the proximal Newton algorithm with multi-stage convex relaxation based on the difference of convex (DC) programming, and enjoys both strong computational and statistical guarantees. Specifically, by leveraging a sophisticated characterization of sparse modeling structures/assumptions (i.e., local restricted strong convexity and Hessian smoothness), we prove that within each stage of convex relaxation, our proposed algorithm achieves (local) quadratic convergence, and eventually obtains a sparse approximate local optimum with optimal statistical properties after only a few convex relaxations. Numerical experiments are provided to support our theory.

stat.ML↗

arXiv2018

Translating Pro-Drop Languages with Reconstruction Models

Pronouns are frequently omitted in pro-drop languages, such as Chinese, generally leading to significant challenges with respect to the production of complete translations. To date, very little attention has been paid to the dropped pronoun (DP) problem within neural machine translation (NMT). In this work, we propose a novel reconstruction-based approach to alleviating DP translation problems for NMT models. Firstly, DPs within all source sentences are automatically annotated with parallel information extracted from the bilingual training corpus. Next, the annotated source sentence is reconstructed from hidden representations in the NMT model. With auxiliary training objectives, in terms of reconstruction scores, the parameters associated with the NMT model are guided to produce enhanced hidden representations that are encouraged as much as possible to embed annotated DP information. Experimental results on both Chinese-English and Japanese-English dialogue translation tasks show that the proposed approach significantly and consistently improves translation performance over a strong NMT baseline, which is directly built on the training data annotated with DPs.

cs.CL↗

arXiv2017

Learning to Remember Translation History with a Continuous Cache

Existing neural machine translation (NMT) models generally translate sentences in isolation, missing the opportunity to take advantage of document-level information. In this work, we propose to augment NMT models with a very light-weight cache-like memory network, which stores recent hidden representations as translation history. The probability distribution over generated words is updated online depending on the translation history retrieved from the memory, endowing NMT models with the capability to dynamically adapt over time. Experiments on multiple domains with different topics and styles show the effectiveness of the proposed approach with negligible impact on the computational cost.

cs.CL↗

arXiv2017

Gradient Sparsification for Communication-Efficient Distributed Optimization

Modern large scale machine learning applications require stochastic optimization algorithms to be implemented on distributed computational architectures. A key bottleneck is the communication overhead for exchanging information such as stochastic gradients among different workers. In this paper, to reduce the communication cost we propose a convex optimization formulation to minimize the coding length of stochastic gradients. To solve the optimal sparsification efficiently, several simple and fast algorithms are proposed for approximate solution, with theoretical guaranteed for sparseness. Experiments on $\ell_2$ regularized logistic regression, support vector machines, and convolutional neural networks validate our sparsification approaches.

cs.LG↗

arXiv2017

Insight-HXMT observations of the first binary neutron star merger GW170817

Finding the electromagnetic (EM) counterpart of binary compact star merger, especially the binary neutron star (BNS) merger, is critically important for gravitational wave (GW) astronomy, cosmology and fundamental physics. On Aug. 17, 2017, Advanced LIGO and \textit{Fermi}/GBM independently triggered the first BNS merger, GW170817, and its high energy EM counterpart, GRB 170817A, respectively, resulting in a global observation campaign covering gamma-ray, X-ray, UV, optical, IR, radio as well as neutrinos. The High Energy X-ray telescope (HE) onboard \textit{Insight}-HXMT (Hard X-ray Modulation Telescope) is the unique high-energy gamma-ray telescope that monitored the entire GW localization area and especially the optical counterpart (SSS17a/AT2017gfo) with very large collection area ($\sim$1000 cm$^2$) and microsecond time resolution in 0.2-5 MeV. In addition, \textit{Insight}-HXMT quickly implemented a Target of Opportunity (ToO) observation to scan the GW localization area for potential X-ray emission from the GW source. Although it did not detect any significant high energy (0.2-5 MeV) radiation from GW170817, its observation helped to confirm the unexpected weak and soft nature of GRB 170817A. Meanwhile, \textit{Insight}-HXMT/HE provides one of the most stringent constraints (~10$^{-7}$ to 10$^{-6}$ erg/cm$^2$/s) for both GRB170817A and any other possible precursor or extended emissions in 0.2-5 MeV, which help us to better understand the properties of EM radiation from this BNS merger. Therefore the observation of \textit{Insight}-HXMT constitutes an important chapter in the full context of multi-wavelength and multi-messenger observation of this historical GW event.

astro-ph.HE↗