arXiv subjects

Yang Wang

Publications and source records attributed to Yang Wang.

At least 595 records · Page 33Linked to original sources

arXiv2022

VSEGAN: Visual Speech Enhancement Generative Adversarial Network

Speech enhancement is an essential task of improving speech quality in noise scenario. Several state-of-the-art approaches have introduced visual information for speech enhancement,since the visual aspect of speech is essentially unaffected by acoustic environment. This paper proposes a novel frameworkthat involves visual information for speech enhancement, by in-corporating a Generative Adversarial Network (GAN). In par-ticular, the proposed visual speech enhancement GAN consistof two networks trained in adversarial manner, i) a generator that adopts multi-layer feature fusion convolution network to enhance input noisy speech, and ii) a discriminator that attemptsto minimize the discrepancy between the distributions of the clean speech signal and enhanced speech signal. Experiment re-sults demonstrated superior performance of the proposed modelagainst several state-of-the-art

eess.AS↗

arXiv2022

Nearly optimal bounds for the global geometric landscape of phase retrieval

The phase retrieval problem is concerned with recovering an unknown signal $\bf{x} \in \mathbb{R}^n$ from a set of magnitude-only measurements $y_j=|\langle \bf{a}_j,\bf{x} \rangle|, \; j=1,\ldots,m$. A natural least squares formulation can be used to solve this problem efficiently even with random initialization, despite its non-convexity of the loss function. One way to explain this surprising phenomenon is the benign geometric landscape: (1) all local minimizers are global; and (2) the objective function has a negative curvature around each saddle point and local maximizer. In this paper, we show that $m=O(n \log n)$ Gaussian random measurements are sufficient to guarantee the loss function of a commonly used estimator has such benign geometric landscape with high probability. This is a step toward answering the open problem given by Sun-Qu-Wright, in which the authors suggest that $O(n \log n)$ or even $O(n)$ is enough to guarantee the favorable geometric property.

cs.IT↗

arXiv2022

BigDL 2.0: Seamless Scaling of AI Pipelines from Laptops to Distributed Cluster

Most AI projects start with a Python notebook running on a single laptop; however, one usually needs to go through a mountain of pains to scale it to handle larger dataset (for both experimentation and production deployment). These usually entail many manual and error-prone steps for the data scientists to fully take advantage of the available hardware resources (e.g., SIMD instructions, multi-processing, quantization, memory allocation optimization, data partitioning, distributed computing, etc.). To address this challenge, we have open sourced BigDL 2.0 at https://github.com/intel-analytics/BigDL/ under Apache 2.0 license (combining the original BigDL and Analytics Zoo projects); using BigDL 2.0, users can simply build conventional Python notebooks on their laptops (with possible AutoML support), which can then be transparently accelerated on a single node (with up-to 9.6x speedup in our experiments), and seamlessly scaled out to a large cluster (across several hundreds servers in real-world use cases). BigDL 2.0 has already been adopted by many real-world users (such as Mastercard, Burger King, Inspur, etc.) in production.

cs.LG↗

arXiv2022

Test suite effectiveness metric evaluation: what do we know and what should we do?

Comparing test suite effectiveness metrics has always been a research hotspot. However, prior studies have different conclusions or even contradict each other for comparing different test suite effectiveness metrics. The problem we found most troubling to our community is that researchers tend to oversimplify the description of the ground truth they use. For example, a common expression is that "we studied the correlation between real faults and the metric to evaluate (MTE)". However, the meaning of "real faults" is not clear-cut. As a result, there is a need to scrutinize the meaning of "real faults". Without this, it will be half-knowledgeable with the conclusions. To tackle this challenge, we propose a framework ASSENT (evAluating teSt Suite EffectiveNess meTrics) to guide the follow-up research. In nature, ASSENT consists of three fundamental components: ground truth, benchmark test suites, and agreement indicator. First, materialize the ground truth for determining the real order in effectiveness among test suites. Second, generate a set of benchmark test suites and derive their ground truth order in effectiveness. Third, for the benchmark test suites, generate the MTE order in effectiveness by the metric to evaluate (MTE). Finally, calculate the agreement indicator between the two orders. Under ASSENT, we are able to compare the accuracy of different test suite effectiveness metrics. We apply ASSENT to evaluate representative test suite effectiveness metrics, including mutation score metrics and code coverage metrics. Our results show that, based on the real faults, mutation score and subsuming mutation score are the best metrics to quantify test suite effectiveness. Meanwhile, by using mutants instead of real faults, MTEs will be overestimated by more than 20% in values.

cs.SE↗

arXiv2022

An error analysis of generative adversarial networks for learning distributions

This paper studies how well generative adversarial networks (GANs) learn probability distributions from finite samples. Our main results establish the convergence rates of GANs under a collection of integral probability metrics defined through Hölder classes, including the Wasserstein distance as a special case. We also show that GANs are able to adaptively learn data distributions with low-dimensional structures or have Hölder densities, when the network architectures are chosen properly. In particular, for distributions concentrated around a low-dimensional set, we show that the learning rates of GANs do not depend on the high ambient dimension, but on the lower intrinsic dimension. Our analysis is based on a new oracle inequality decomposing the estimation error into the generator and discriminator approximation error and the statistical error, which may be of independent interest.

cs.LG↗

arXiv2022

Ab initio Approaches to High Entropy Alloys: A Comparison of CPA, SQS, and Supercell Methods

We present a comparative study of different modeling approaches to the electronic properties of the $\textrm{Hf}_{0.05}\textrm{Nb}_{0.05}\textrm{Ta}_{0.8}\textrm{Ti}_{0.05}\textrm{Zr}_{0.05}$ high entropy alloy. Common to our modeling is the methodology to compute the one-particle Green's function in the framework of density functional theory. We demonstrate that the special quasi-random structures modeling and the supercell, i.e. the locally self-consistent multiple-scatering methods provide very similar results for the ground state properties such as the spectral function (density of states) and the equilibrium lattice parameter. To reconcile the multiple-scattering single-site coherent potential approximation with the real space supercell methods, we included the effect of screening of the net charges of the alloy components. Based on the analysis of the total energy and spectral functions computed within the density functional theory, we found no signature for the long-range or local magnetic moments formation in the $\textrm{Hf}_{0.05}\textrm{Nb}_{0.05}\textrm{Ta}_{0.8}\textrm{Ti}_{0.05}\textrm{Zr}_{0.05}$ high entropy alloy, instead we find possible superconductivity below $\sim 9$K.

cond-mat.mtrl-sci↗

arXiv2022

ProgressiveMotionSeg: Mutually Reinforced Framework for Event-Based Motion Segmentation

Dynamic Vision Sensor (DVS) can asynchronously output the events reflecting apparent motion of objects with microsecond resolution, and shows great application potential in monitoring and other fields. However, the output event stream of existing DVS inevitably contains background activity noise (BA noise) due to dark current and junction leakage current, which will affect the temporal correlation of objects, resulting in deteriorated motion estimation performance. Particularly, the existing filter-based denoising methods cannot be directly applied to suppress the noise in event stream, since there is no spatial correlation. To address this issue, this paper presents a novel progressive framework, in which a Motion Estimation (ME) module and an Event Denoising (ED) module are jointly optimized in a mutually reinforced manner. Specifically, based on the maximum sharpness criterion, ME module divides the input event into several segments by adaptive clustering in a motion compensating warp field, and captures the temporal correlation of event stream according to the clustered motion parameters. Taking temporal correlation as guidance, ED module calculates the confidence that each event belongs to real activity events, and transmits it to ME module to update energy function of motion segmentation for noise suppression. The two steps are iteratively updated until stable motion segmentation results are obtained. Extensive experimental results on both synthetic and real datasets demonstrate the superiority of our proposed approaches against the State-Of-The-Art (SOTA) methods.

cs.CV↗

arXiv2022

Revealing room temperature ferromagnetism in exfoliated Fe$_5$GeTe$_2$ flakes with quantum magnetic imaging

Van der Waals material Fe$_5$GeTe$_2$, with its long-range ferromagnetic ordering near room temperature, has significant potential to become an enabling platform for implementing novel spintronic and quantum devices. To pave the way for applications, it is crucial to determine the magnetic properties when the thickness of Fe5GeTe2 reaches the few-layers regime. However, this is highly challenging due to the need for a characterization technique that is local, highly sensitive, artifact-free, and operational with minimal fabrication. Prior studies have indicated that Curie temperature TC can reach up to close to room temperature for exfoliated Fe$_5$GeTe$_2$ flakes, as measured via electrical transport; there is a need to validate these results with a measurement that reveals magnetism more directly. In this work, we investigate the magnetic properties of exfoliated thin flakes of van der Waals magnet Fe$_5$GeTe$_2$ via a quantum magnetic imaging technique based on nitrogen vacancy diamond. Through imaging the stray fields, we confirm room-temperature magnetic order in Fe$_5$GeTe$_2$ thin flakes with thickness down to 7 units cell. The stray field patterns and their response to magnetizing fields with different polarities point to a perpendicular easy-axis anisotropy. Furthermore, we perform imaging at different temperatures and determine the Curie temperature of the flakes at Tc~300 K. These results provide the basis for realizing a room-temperature monolayer ferromagnet with Fe$_5$GeTe$_2$. This work also demonstrates that the imaging technique enables a rapid screening of multiple flakes simultaneously, thereby paving the way towards high throughput characterization of potential 2D magnets near room temperature.

cond-mat.mes-hall↗

arXiv2022

A rate of convergence of Physics Informed Neural Networks for the linear second order elliptic PDEs

In recent years, physical informed neural networks (PINNs) have been shown to be a powerful tool for solving PDEs empirically. However, numerical analysis of PINNs is still missing. In this paper, we prove the convergence rate to PINNs for the second order elliptic equations with Dirichlet boundary condition, by establishing the upper bounds on the number of training samples, depth and width of the deep neural networks to achieve desired accuracy. The error of PINNs is decomposed into approximation error and statistical error, where the approximation error is given in $C^2$ norm with $\mathrm{ReLU}^{3}$ networks (deep network with activations function $\max\{0,x^3\}$) and the statistical error is estimated by Rademacher complexity. We derive the bound on the Rademacher complexity of the non-Lipschitz composition of gradient norm with $\mathrm{ReLU}^{3}$ network, which is of immense independent interest.

math.NA↗

arXiv2022

Sensitivity tests of cosmic velocity fields to massive neutrinos

We investigate impacts of massive neutrinos on the cosmic velocity fields, employing high-resolution cosmological $N$-body simulations provided by the information-optimized CUBE code, where cosmic neutrinos are evolved using collisionless hydrodynamics and their perturbations can be accurately resolved. In this study we focus, for the first time, on the analysis of massive-neutrino induced suppression effects in various cosmic velocity field components of velocity magnitude, divergence, vorticity and dispersion. By varying the neutrino mass sum $M_ν$ from 0 -- 0.4 eV, the simulations show that, the power spectra of vorticity -- exclusively sourced by non-linear structure formation that is affected by massive neutrinos significantly -- is very sensitive to the mass sum, which potentially provide novel signatures in detecting massive neutrinos. Furthermore, using the chi-square statistic, we quantitatively test the sensitivity of the density and velocity power spectra to the neutrino mass sum. Indeed, we find that, the vorticity spectrum has the highest sensitivity, and the null hypothesis of massless neutrinos is incompatible with both vorticity and divergence spectra from $M_ν=0.1$ eV at high significance ($p$-value $= 0.03$ and $0.07$, respectively). These results demonstrate clearly the importance of peculiar velocity field measurements, in particular of vorticity and divergence components, in determination of neutrino mass and mass hierarchy.

astro-ph.CO↗

arXiv2022

When Does A Spectral Graph Neural Network Fail in Node Classification?

Spectral Graph Neural Networks (GNNs) with various graph filters have received extensive affirmation due to their promising performance in graph learning problems. However, it is known that GNNs do not always perform well. Although graph filters provide theoretical foundations for model explanations, it is unclear when a spectral GNN will fail. In this paper, focusing on node classification problems, we conduct a theoretical analysis of spectral GNNs performance by investigating their prediction error. With the aid of graph indicators including homophily degree and response efficiency we proposed, we establish a comprehensive understanding of complex relationships between graph structure, node labels, and graph filters. We indicate that graph filters with low response efficiency on label difference are prone to fail. To enhance GNNs performance, we provide a provably better strategy for filter design from our theoretical analysis - using data-driven filter banks, and propose simple models for empirical validation. Experimental results show consistency with our theoretical results and support our strategy.

cs.LG↗

arXiv2022

BiGCN: A Bi-directional Low-Pass Filtering Graph Neural Network

Graph convolutional networks (GCNs) have achieved great success on graph-structured data. Many graph convolutional networks can be thought of as low-pass filters for graph signals. In this paper, we propose a more powerful graph convolutional network, named BiGCN, that extends to bidirectional filtering. Specifically, we not only consider the original graph structure information but also the latent correlation between features, thus BiGCN can filter the signals along with both the original graph and a latent feature-connection graph. Compared with most existing GCNs, BiGCN is more robust and has powerful capacities for feature denoising. We perform node classification and link prediction in citation networks and co-purchase networks with three settings: noise-rate, noise-level and structure-mistakes. Extensive experimental results demonstrate that our model outperforms the state-of-the-art graph neural networks in both clean and artificially noisy data.

cs.LG↗

arXiv2022

Wasserstein Graph Neural Networks for Graphs with Missing Attributes

Missing node attributes is a common problem in real-world graphs. Graph neural networks have been demonstrated power in graph representation learning while their performance is affected by the completeness of graph information. Most of them are not specified for missing-attribute graphs and fail to leverage incomplete attribute information effectively. In this paper, we propose an innovative node representation learning framework, Wasserstein Graph Neural Network (WGNN), to mitigate the problem. To make the most of limited observed attribute information and capture the uncertainty caused by missing values, we express nodes as low-dimensional distributions derived from the decomposition of the attribute matrix. Furthermore, we strengthen the expressiveness of representations by developing a novel message passing schema that aggregates distributional information from neighbors in the Wasserstein space. We test WGNN in node classification tasks under two missing-attribute cases on both synthetic and real-world datasets. In addition, we find WGNN suitable to recover missing values and adapt them to tackle matrix completion problems with graphs of users and items. Experimental results on both tasks demonstrate the superiority of our method.

cs.LG↗

arXiv2022

Near-optimal Top-k Pattern Mining

Nowadays, frequent pattern mining (FPM) on large graphs receives increasing attention, since it is crucial to a variety of applications, e.g., social analysis. Informally, the FPM problem is defined as finding all the patterns in a large graph with frequency above a user-defined threshold. However, this problem is nontrivial due to the unaffordable computational and space costs in the mining process. In light of this, we propose a cost-effective approach to mining near-optimal top-k patterns. Our approach applies a "level-wise" strategy to incrementally detect frequent patterns, hence is able to terminate as soon as top-k patterns are discovered. Moreover, we develop a technique to compute the lower bound of support with smart traverse strategy and compact data structures. Extensive experimental studies on real-life and synthetic graphs show that our approach performs well, i.e., it outperforms traditional counterparts in efficiency, memory footprint, recall and scalability.

cs.DB↗

arXiv2022

DECORE: Deep Compression with Reinforcement Learning

Deep learning has become an increasingly popular and powerful methodology for modern pattern recognition systems. However, many deep neural networks have millions or billions of parameters, making them untenable for real-world applications due to constraints on memory size or latency requirements. As a result, efficient network compression techniques are often required for the widespread adoption of deep learning methods. We present DECORE, a reinforcement learning-based approach to automate the network compression process. DECORE assigns an agent to each channel in the network along with a light policy gradient method to learn which neurons or channels to be kept or removed. Each agent in the network has just one parameter (keep or drop) to learn, which leads to a much faster training process compared to existing approaches. DECORE provides state-of-the-art compression results on various network architectures and various datasets. For example, on the ResNet-110 architecture, it achieves a 64.8% compression and 61.8% FLOPs reduction as compared to the baseline model without any accuracy loss on the CIFAR-10 dataset. It can reduce the size of regular architectures like the VGG network by up to 99% with just a small accuracy drop of 2.28%. For a larger dataset like ImageNet with just 30 epochs of training, it can compress the ResNet-50 architecture by 44.7% and reduce FLOPs by 42.3%, with just a 0.69% drop on Top-5 accuracy of the uncompressed model. We also demonstrate that DECORE can be used to search for compressed network architectures based on various constraints, such as memory and FLOPs.

cs.LG↗

arXiv2022

Spatio-Temporal Self-Attention Network for Video Saliency Prediction

3D convolutional neural networks have achieved promising results for video tasks in computer vision, including video saliency prediction that is explored in this paper. However, 3D convolution encodes visual representation merely on fixed local spacetime according to its kernel size, while human attention is always attracted by relational visual features at different time. To overcome this limitation, we propose a novel Spatio-Temporal Self-Attention 3D Network (STSANet) for video saliency prediction, in which multiple Spatio-Temporal Self-Attention (STSA) modules are employed at different levels of 3D convolutional backbone to directly capture long-range relations between spatio-temporal features of different time steps. Besides, we propose an Attentional Multi-Scale Fusion (AMSF) module to integrate multi-level features with the perception of context in semantic and spatio-temporal subspaces. Extensive experiments demonstrate the contributions of key components of our method, and the results on DHF1K, Hollywood-2, UCF, and DIEM benchmark datasets clearly prove the superiority of the proposed model compared with all state-of-the-art models.

cs.CV↗

arXiv2022

"It's A Blessing and A Curse": Unpacking Creators' Practices with Non-Fungible Tokens (NFTs) and Their Communities

NFTs (Non-Fungible Tokens) are blockchain-based cryptographic tokens to represent ownership of unique content such as images, videos, or 3D objects. Despite NFTs' increasing popularity and skyrocketing trading prices, little is known about people's perceptions of and experiences with NFTs. In this work, we focus on NFT creators and present results of an exploratory qualitative study in which we interviewed 15 NFT creators from nine different countries. Our participants had nuanced feelings about NFTs and their communities. We found that most of our participants were enthusiastic about the underlying technologies and how they empower individuals to express their creativity and pursue new business models of content creation. Our participants also gave kudos to the NFT communities that have supported them to learn, collaborate, and grow in their NFT endeavors. However, these positivities were juxtaposed by their accounts of the many challenges that they encountered and thorny issues that the NFT ecosystem is grappling with around ownership of digital content, low-quality NFTs, scams, possible money laundering, and regulations. We discuss how the built-in properties (e.g., decentralization) of blockchains and NFTs might have contributed to some of these issues. We present design implications on how to improve the NFT ecosystem (e.g., making NFTs even more accessible to newcomers and the broader population).

cs.CY↗

arXiv2022

Self-supervised Spatiotemporal Representation Learning by Exploiting Video Continuity

Recent self-supervised video representation learning methods have found significant success by exploring essential properties of videos, e.g. speed, temporal order, etc. This work exploits an essential yet under-explored property of videos, the video continuity, to obtain supervision signals for self-supervised representation learning. Specifically, we formulate three novel continuity-related pretext tasks, i.e. continuity justification, discontinuity localization, and missing section approximation, that jointly supervise a shared backbone for video representation learning. This self-supervision approach, termed as Continuity Perception Network (CPNet), solves the three tasks altogether and encourages the backbone network to learn local and long-ranged motion and context representations. It outperforms prior arts on multiple downstream tasks, such as action recognition, video retrieval, and action localization. Additionally, the video continuity can be complementary to other coarse-grained video properties for representation learning, and integrating the proposed pretext task to prior arts can yield much performance gains.

cs.CV↗