arXiv subjects

Chao Zhang

Publications and source records attributed to Chao Zhang.

At least 1,027 records · Page 57Linked to original sources

arXiv2019

Recent Advances on HEVC Inter-frame Coding: From Optimization to Implementation and Beyond

High Efficiency Video Coding (HEVC) has doubled the video compression ratio with equivalent subjective quality as compared to its predecessor H.264/AVC. The significant coding efficiency improvement is attributed to many new techniques. Inter-frame coding is one of the most powerful yet complicated techniques therein and has posed high computational burden thus main obstacle in HEVC-based real-time applications. Recently, plenty of research has been done to optimize the inter-frame coding, either to reduce the complexity for real-time applications, or to further enhance the encoding efficiency. In this paper, we provide a comprehensive review of the state-of-the-art techniques for HEVC inter-frame coding from three aspects, namely fast inter coding solutions, implementation on different hardware platforms as well as advanced inter coding techniques. More specifically, different algorithms in each aspect are further subdivided into sub-categories and compared in terms of pros, cons, coding efficiency and coding complexity. To the best of our knowledge, this is the first such comprehensive review of the recent advances of the inter-frame coding for HEVC and hopefully it would help the improvement, implementation and applications of HEVC as well as the ongoing development of the next generation video coding standard.

eess.IV↗

arXiv2019

Temperature effects on the ionic conductivity in concentrated alkaline electrolyte solutions

Alkaline electrolyte solutions are important components in rechargeable batteries and alkaline fuel cells. As the ionic conductivity is thought to be a limiting factor in the performance of these devices, which are often operated at elevated temperatures, its temperature dependence is of significant interest. Here we use NaOH as a prototypical example of alkaline electrolytes, and for this system we have carried out reactive molecular dynamics simulations with an experimentally verified high-dimensional neural network potential derived from density-functional theory calculations. It is found that in concentrated NaOH solutions elevated temperatures enhance both the contributions from proton transfer to the ionic conductivity and deviations from the Nernst-Einstein relation. These findings are expected to be of practical relevance for electrochemical devices based on alkaline electrolyte solutions.

physics.chem-ph↗

arXiv2019

Energy-Efficient MIMO Multiuser Systems: Nash Equilibrium Analysis

In this paper, an energy efficiency (EE) game in a MIMO multiple access channel (MAC) communication system is considered. The existence and the uniqueness of the Nash Equilibrium (NE) is affirmed. A bisection search algorithm is designed to find this unique NE. Despite being sub-optimal for deploying the $\varepsilon$-approximate NE of the game when the number of antennas in transmitter is unequal to receiver's, the policy found by the proposed algorithm is shown to be more efficient than the classical allocation techniques. Moreover, compared to the general algorithm based on fractional programming technique, our proposed algorithm is easier to implement. Simulation shows that even the policy found by proposed algorithm is not the NE of the game, the deviation w.r.t. to the exact NE is small and the resulted policy actually Pareto-dominates the unique NE of the game at least for 2-user situation.

eess.SP↗

arXiv2019

Sparse DNNs with Improved Adversarial Robustness

Deep neural networks (DNNs) are computationally/memory-intensive and vulnerable to adversarial attacks, making them prohibitive in some real-world applications. By converting dense models into sparse ones, pruning appears to be a promising solution to reducing the computation/memory cost. This paper studies classification models, especially DNN-based ones, to demonstrate that there exists intrinsic relationships between their sparsity and adversarial robustness. Our analyses reveal, both theoretically and empirically, that nonlinear DNN-based classifiers behave differently under $l_2$ attacks from some linear ones. We further demonstrate that an appropriately higher model sparsity implies better robustness of nonlinear DNNs, whereas over-sparsified models can be more difficult to resist adversarial examples.

cs.LG↗

arXiv2019

Spherical Text Embedding

Unsupervised text embedding has shown great power in a wide range of NLP tasks. While text embeddings are typically learned in the Euclidean space, directional similarity is often more effective in tasks such as word similarity and document clustering, which creates a gap between the training stage and usage stage of text embedding. To close this gap, we propose a spherical generative model based on which unsupervised word and paragraph embeddings are jointly learned. To learn text embeddings in the spherical space, we develop an efficient optimization algorithm with convergence guarantee based on Riemannian optimization. Our model enjoys high efficiency and achieves state-of-the-art performances on various text embedding tasks including word similarity and document clustering.

cs.CL↗

arXiv2019

Demonstrating quantum coherence and metrology that is resilient to transversal noise

Quantum systems can be exploited for disruptive technologies but in practice quantum features are fragile due to noisy environments. Quantum coherence, a fundamental such feature, is a basis-dependent property that is known to exhibit a resilience to certain types of Markovian noise. Yet, it is still unclear whether this resilience can be relevant in practical tasks. Here, we experimentally investigate the resilient effect of quantum coherence in a photonic Greenberger-Horne-Zeilinger state under Markovian bit-flip noise, and explore its applications in a noisy metrology scenario. In particular, using up to six-qubit probes, we demonstrate that the standard quantum limit can be outperformed under a transversal noise strength of approximately equal magnitude to the signal, providing experimental evidence of metrological advantage even in the presence of uncorrelated Markovian noise. This work highlights the important role of passive control in noisy quantum hardware, which can act as a low-overhead complement to more traditional approaches such as quantum error correction, thus impacting on the deployment of quantum technologies in real-world settings.

quant-ph↗

arXiv2019

Distribution of high-dimensional orbital angular momentum entanglement at telecom wavelength over 1km of optical fibre

High-dimensional entanglement has demonstrated potential for increasing channel capacity and resistance to noise in quantum information processing. However, its distribution is a challenging task, imposing a severe restriction on its application. Here we report the first distribution of three-dimensional orbital angular momentum (OAM) entanglement via a 1-km-long optical fibre. Using an actively-stabilizing phase pre-compensation technique, we successfully transport one photon of a three-dimensional OAM entangled photon pair through the fibre. The distributed OAM entangled state still shows a fidelity up to 71\% with respect to the three-dimensional maximal-entangled-state (MES). In addition, we certify that the high-dimensional quantum entanglement survives the transportation by violating a generalized Bell inequality, obtaining a violation of $\sim3$ standard deviations with $I_{3}=2.12\pm0.04$. The method we developed can be extended to higher OAM dimension and larger distances in principle. Our results make a significant step towards future OAM-based high-dimensional long-distance quantum communication.

quant-ph↗

arXiv2019

Efficient Projection-Free Online Methods with Stochastic Recursive Gradient

This paper focuses on projection-free methods for solving smooth Online Convex Optimization (OCO) problems. Existing projection-free methods either achieve suboptimal regret bounds or have high per-iteration computational costs. To fill this gap, two efficient projection-free online methods called ORGFW and MORGFW are proposed for solving stochastic and adversarial OCO problems, respectively. By employing a recursive gradient estimator, our methods achieve optimal regret bounds (up to a logarithmic factor) while possessing low per-iteration computational costs. Experimental results demonstrate the efficiency of the proposed methods compared to state-of-the-arts.

cs.LG↗

arXiv2019

Beyond Human Parts: Dual Part-Aligned Representations for Person Re-Identification

Person re-identification is a challenging task due to various complex factors. Recent studies have attempted to integrate human parsing results or externally defined attributes to help capture human parts or important object regions. On the other hand, there still exist many useful contextual cues that do not fall into the scope of predefined human parts or attributes. In this paper, we address the missed contextual cues by exploiting both the accurate human parts and the coarse non-human parts. In our implementation, we apply a human parsing model to extract the binary human part masks \emph{and} a self-attention mechanism to capture the soft latent (non-human) part masks. We verify the effectiveness of our approach with new state-of-the-art performances on three challenging benchmarks: Market-1501, DukeMTMC-reID and CUHK03. Our implementation is available at https://github.com/ggjy/P2Net.pytorch.

cs.CV↗

arXiv2019

Aggregated Gradient Langevin Dynamics

In this paper, we explore a general Aggregated Gradient Langevin Dynamics framework (AGLD) for the Markov Chain Monte Carlo (MCMC) sampling. We investigate the nonasymptotic convergence of AGLD with a unified analysis for different data accessing (e.g. random access, cyclic access and random reshuffle) and snapshot updating strategies, under convex and nonconvex settings respectively. It is the first time that bounds for I/O friendly strategies such as cyclic access and random reshuffle have been established in the MCMC literature. The theoretic results also indicate that methods in AGLD possess the merits of both the low per-iteration computational complexity and the short mixture time. Empirical studies demonstrate that our framework allows to derive novel schemes to generate high-quality samples for large-scale Bayesian posterior learning tasks.

cs.LG↗

arXiv2019

HiExpan: Task-Guided Taxonomy Construction by Hierarchical Tree Expansion

Taxonomies are of great value to many knowledge-rich applications. As the manual taxonomy curation costs enormous human effects, automatic taxonomy construction is in great demand. However, most existing automatic taxonomy construction methods can only build hypernymy taxonomies wherein each edge is limited to expressing the "is-a" relation. Such a restriction limits their applicability to more diverse real-world tasks where the parent-child may carry different relations. In this paper, we aim to construct a task-guided taxonomy from a domain-specific corpus and allow users to input a "seed" taxonomy, serving as the task guidance. We propose an expansion-based taxonomy construction framework, namely HiExpan, which automatically generates key term list from the corpus and iteratively grows the seed taxonomy. Specifically, HiExpan views all children under each taxonomy node forming a coherent set and builds the taxonomy by recursively expanding all these sets. Furthermore, HiExpan incorporates a weakly-supervised relation extraction module to extract the initial children of a newly-expanded node and adjusts the taxonomy tree by optimizing its global structure. Our experiments on three real datasets from different domains demonstrate the effectiveness of HiExpan for building task-guided taxonomies.

cs.CL↗

arXiv2019

The effect of disorder on the phase diagrams of hard-core lattice bosons with cavity-mediated long-range and nearest-neighbor interactions

We use quantum Monte Carlo simulations with the worm algorithm to study the phase diagram of a two-dimensional Bose-Hubbard model with cavity-mediated long-range interactions and uncorrelated disorder in the hard-core limit. Our study shows the system is in a supersolid phase at weak disorder and a disordered solid phase at stronger disorder. Due to long-range interactions, a large region of metastable states exists in both clean and disordered systems. By comparing the phase diagrams for both clean and disordered systems, we find that disorder suppresses metastable states and superfluidity. We compare these results with the phase diagram of the extended Bose-Hubbard model with nearest-neighbor interactions. Here, the supersolid phase does not exist even at weak disorder. We identify two kinds of glassy phases: a Bose glass phase and a disordered solid phase. The glassy phases intervene between the density-wave and superfluid phases as the Griffiths phase of the Bose-Hubbard model. The disordered solid phase intervenes between the density-wave and Bose glass phases since both have a finite structure factor.

cond-mat.dis-nn↗

arXiv2019

On Dimension-free Tail Inequalities for Sums of Random Matrices and Applications

In this paper, we present a new framework to obtain tail inequalities for sums of random matrices. Compared with existing works, our tail inequalities have the following characteristics: 1) high feasibility--they can be used to study the tail behavior of various matrix functions, e.g., arbitrary matrix norms, the absolute value of the sum of the sum of the $j$ largest singular values (resp. eigenvalues) of complex matrices (resp. Hermitian matrices); and 2) independence of matrix dimension --- they do not have the matrix-dimension term as a product factor, and thus are suitable to the scenario of high-dimensional or infinite-dimensional random matrices. The price we pay to obtain these advantages is that the convergence rate of the resulting inequalities will become slow when the number of summand random matrices is large. We also develop the tail inequalities for matrix random series and matrix martingale difference sequence. We also demonstrate usefulness of our tail bounds in several fields. In compressed sensing, we employ the resulted tail inequalities to achieve a proof of the restricted isometry property when the measurement matrix is the sum of random matrices without any assumption on the distributions of matrix entries. In probability theory, we derive a new upper bound to the supreme of stochastic processes. In machine learning, we prove new expectation bounds of sums of random matrices matrix and obtain matrix approximation schemes via random sampling. In quantum information, we show a new analysis relating to the fractional cover number of quantum hypergraphs. In theoretical computer science, we obtain randomness-efficient samplers using matrix expander graphs that can be efficiently implemented in time without dependence on matrix dimensions.

cs.LG↗

arXiv2019

Gravitational waveforms and radiation powers of the triple system PSR J0337+1715 in modified theories of gravity

In this paper, we study the gravitational waveforms, polarizations and radiation powers of the first relativistic triple systems PSR J0337 + 1715, observed in 2014, by using the post-Newtonian approximations to their lowest order. Although they cannot be observed either by current or next generation of the detectors, they do provide useful information to test different theories of gravity. In particular, we carry out the studies in three different theories, general relativity (GR), Brans-Dicke (BD) gravity, and Einstein-aether ($æ$) theory. The tensor modes $h_{+}$ and $h_{\times}$ exist in all three theories and have almost equal amplitudes. Their frequencies are all peaked at two locations, $ f^{(+, \times)}_1 \simeq 0.0686656 μ$Hz and $ f^{(+, \times)}_2 \simeq 14.2138 μ$Hz, which are about twice of the outer and inner orbital frequencies of the triple system, as predicted in GR. In $æ$-theory, all the six polarization modes are different from zero, but the breathing $h_b$ and longitudinal $h_L$ modes are not independent and also peaked at two frequencies, but at the frequencies, $ f^{(b, L, æ)}_{1} \simeq 0.0457771 μ$Hz and $f^{(b, L, æ)}_{ 2} \simeq 7.09545 μ$Hz. A similar phenomenon is also observed in BD gravity, in which only the three modes $h_{+},\; h_{\times}$ and $h_{b}$ exit. We also study the radiation powers, and find that the quadrupole emission in each of the three theories has almost the same amplitude, but the dipole emission can be as big as the quadrupole emission in $æ$-theory. This can provide a very promising window to obtain severe constraints on $æ$-theory by the multi-band gravitational wave astronomy.

astro-ph.HE↗

arXiv2019

Multi-Span Acoustic Modelling using Raw Waveform Signals

Traditional automatic speech recognition (ASR) systems often use an acoustic model (AM) built on handcrafted acoustic features, such as log Mel-filter bank (FBANK) values. Recent studies found that AMs with convolutional neural networks (CNNs) can directly use the raw waveform signal as input. Given sufficient training data, these AMs can yield a competitive word error rate (WER) to those built on FBANK features. This paper proposes a novel multi-span structure for acoustic modelling based on the raw waveform with multiple streams of CNN input layers, each processing a different span of the raw waveform signal. Evaluation on both the single channel CHiME4 and AMI data sets show that multi-span AMs give a lower WER than FBANK AMs by an average of about 5% (relative). Analysis of the trained multi-span model reveals that the CNNs can learn filters that are rather different to the log Mel filters. Furthermore, the paper shows that a widely used single span raw waveform AM can be improved by using a smaller CNN kernel size and increased stride to yield improved WERs.

eess.AS↗

arXiv2019

Integrating Source-channel and Attention-based Sequence-to-sequence Models for Speech Recognition

This paper proposes a novel automatic speech recognition (ASR) framework called Integrated Source-Channel and Attention (ISCA) that combines the advantages of traditional systems based on the noisy source-channel model (SC) and end-to-end style systems using attention-based sequence-to-sequence models. The traditional SC system framework includes hidden Markov models and connectionist temporal classification (CTC) based acoustic models, language models (LMs), and a decoding procedure based on a lexicon, whereas the end-to-end style attention-based system jointly models the whole process with a single model. By rescoring the hypotheses produced by traditional systems using end-to-end style systems based on an extended noisy source-channel model, ISCA allows structured knowledge to be easily incorporated via the SC-based model while exploiting the complementarity of the attention-based model. Experiments on the AMI meeting corpus show that ISCA is able to give a relative word error rate reduction up to 21% over an individual system, and by 13% over an alternative method which also involves combining CTC and attention-based models.

eess.AS↗

arXiv2019

CubeNet: Multi-Facet Hierarchical Heterogeneous Network Construction, Analysis, and Mining

Due to the ever-increasing size of data, construction, analysis and mining of universal massive networks are becoming forbidden and meaningless. In this work, we outline a novel framework called CubeNet, which systematically constructs and organizes real-world networks into different but correlated semantic cells, to support various downstream network analysis and mining tasks with better flexibility, deeper insights and higher efficiency. Particular, we promote our recent research on text and network mining with novel concepts and techniques to (1) construct four real-world large-scale multi-facet hierarchical heterogeneous networks; (2) enable insightful OLAP-style network analysis; (3) facilitate localized and contextual network mining. Although some functions have been covered individually in our previous work, a systematic and efficient realization of an organic system has not been studied, while some functions are still our on-going research tasks. By integrating them, CubeNet may not only showcase the utility of our recent research, but also inspire and stimulate future research on effective, insightful and scalable knowledge discovery under this novel framework.

cs.SI↗

arXiv2019

Decision Set Optimization and Energy-Efficient MIMO Communications

Assuming that the number of possible decisions for a transmitter (e.g., the number of possible beamforming vectors) has to be finite and is given, this paper investigates for the first time the problem of determining the best decision set when energy-efficiency maximization is pursued. We propose a framework to find a good (finite) decision set which induces a minimal performance loss w.r.t. to the continuous case. We exploit this framework for a scenario of energy-efficient MIMO communications in which transmit power and beamforming vectors have to be adapted jointly to the channel given under finite-rate feedback. To determine a good decision set we propose an algorithm which combines the approach of Invasive Weed Optimization (IWO) and an Evolutionary Algorithm (EA). We provide a numerical analysis which illustrates the benefits of our point of view. In particular, given a performance loss level, the feedback rate can by reduced by 2 when the transmit decision set has been designed properly by using our algorithm. The impact on energy-efficiency is also seen to be significant.

cs.IT↗