arXiv Science⌕ Search

arXiv subjects

Zhiyong Wang

Publications and source records attributed to Zhiyong Wang.

At least 181 records · Page 10Linked to original sources
arXiv2021

Sign Language Translation with Hierarchical Spatio-TemporalGraph Neural Network

Sign language translation (SLT), which generates text in a spoken language from visual content in a sign language, is important to assist the hard-of-hearing community for their communications. Inspired by neural machine translation (NMT), most existing SLT studies adopted a general sequence to sequence learning strategy. However, SLT is significantly different from general NMT tasks since sign languages convey messages through multiple visual-manual aspects. Therefore, in this paper, these unique characteristics of sign languages are formulated as hierarchical spatio-temporal graph representations, including high-level and fine-level graphs of which a vertex characterizes a specified body part and an edge represents their interactions. Particularly, high-level graphs represent the patterns in the regions such as hands and face, and fine-level graphs consider the joints of hands and landmarks of facial regions. To learn these graph patterns, a novel deep learning architecture, namely hierarchical spatio-temporal graph neural network (HST-GNN), is proposed. Graph convolutions and graph self-attentions with neighborhood context are proposed to characterize both the local and the global graph properties. Experimental results on benchmark datasets demonstrated the effectiveness of the proposed method.

cs.CV↗
arXiv2021

An overview of $η_c(1S)$, $η_c(2S)$ and $h_c(1P)$ physics at BESIII

With the help of the largest data samples of $J/ψ$ and $ψ(2S)$ events ever produced in $e^+e^-$ annihilations, the three singlet charmonium states, $η_c(1S)$, $η_c(2S)$ and $h_c(1P)$, have been extensively studied at the BESIII experiment. In this review, a survey on the most recent results, including a series of precision measurements and observations of their new decay modes, is presented, which indicates the further investigations on their decays are needed to understand their decay mechanisms and have precision tests of the theoretical models. At present, about eight times larger data samples of 10 billion $J/ψ$ events and 3 billion $ψ(3686)$ events were collected with the BESIII detector, and thus the prospects for the study of these three charmonium states is discussed extensively.

hep-ex↗
arXiv2021

Boundedness of operators generated by fractional semigroups associated with Schrödinger operators on Campanato type spaces via $T1$ theorem

Let $\mathcal{L}=-Δ+V$ be a Schrödinger operator, where the nonnegative potential $V$ belongs to the reverse Hölder class $B_{q}$. By the aid of the subordinative formula, we estimate the regularities of the fractional heat semigroup, $\{e^{-t\mathcal{L}^α}\}_{t>0},$ associated with $\mathcal{L}$. As an application, we obtain the $BMO^γ_{\mathcal{L}}$-boundedness of the maximal function, and the Littlewood-Paley $g$-functions associated with $\mathcal{L}$ via $T1$ theorem, respectively.

math.CA↗
arXiv2021

Observation of superluminal signaling of terahertz pulses

Superluminal tunneling of light through a barrier has attracted broad interest in the last several decades. Despite the observation of such phenomena in various systems, it has been under intensive debate whether the transmitted light truly carry the information of the original pulse. Here we report observation of anomalous time response for terahertz electromagnetic pulses passing through thin metal films, with the pulse shape of the transmitted beam faithfully resembling that of the incident beam. A causal theoretical analysis is developed to explain the experiments, though the theory of Special Relativity may confront a challenge in this exceptional circumstance. These findings may facilitate future applications in high-speed optical communication or signal transmission, and may reshape our fundamental understanding about the tunneling of light.

physics.optics↗
arXiv2020

Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition

Spatial-temporal graphs have been widely used by skeleton-based action recognition algorithms to model human action dynamics. To capture robust movement patterns from these graphs, long-range and multi-scale context aggregation and spatial-temporal dependency modeling are critical aspects of a powerful feature extractor. However, existing methods have limitations in achieving (1) unbiased long-range joint relationship modeling under multi-scale operators and (2) unobstructed cross-spacetime information flow for capturing complex spatial-temporal dependencies. In this work, we present (1) a simple method to disentangle multi-scale graph convolutions and (2) a unified spatial-temporal graph convolutional operator named G3D. The proposed multi-scale aggregation scheme disentangles the importance of nodes in different neighborhoods for effective long-range modeling. The proposed G3D module leverages dense cross-spacetime edges as skip connections for direct information propagation across the spatial-temporal graph. By coupling these proposals, we develop a powerful feature extractor named MS-G3D based on which our model outperforms previous state-of-the-art methods on three large-scale datasets: NTU RGB+D 60, NTU RGB+D 120, and Kinetics Skeleton 400.

cs.CV↗
arXiv2019

Coupling Matrix Manifolds and Their Applications in Optimal Transport

Optimal transport (OT) is a powerful tool for measuring the distance between two defined probability distributions. In this paper, we develop a new manifold named the coupling matrix manifold (CMM), where each point on CMM can be regarded as the transportation plan of the OT problem. We firstly explore the Riemannian geometry of CMM with the metric expressed by the Fisher information. These geometrical features of CMM have paved the way for developing numerical Riemannian optimization algorithms such as Riemannian gradient descent and Riemannian trust-region algorithms, forming a uniform optimization method for all types of OT problems. The proposed method is then applied to solve several OT problems studied by previous literature. The results of the numerical experiments illustrate that the optimization algorithms that are based on the method proposed in this paper are comparable to the classic ones, for example, the Sinkhorn algorithm, while outperforming other state-of-the-art algorithms without considering the geometry information, especially in the case of non-entropy optimal transport.

cs.LG↗
arXiv2019

IntersectGAN: Learning Domain Intersection for Generating Images with Multiple Attributes

Generative adversarial networks (GANs) have demonstrated great success in generating various visual content. However, images generated by existing GANs are often of attributes (e.g., smiling expression) learned from one image domain. As a result, generating images of multiple attributes requires many real samples possessing multiple attributes which are very resource expensive to be collected. In this paper, we propose a novel GAN, namely IntersectGAN, to learn multiple attributes from different image domains through an intersecting architecture. For example, given two image domains $X_1$ and $X_2$ with certain attributes, the intersection $X_1 \cap X_2$ denotes a new domain where images possess the attributes from both $X_1$ and $X_2$ domains. The proposed IntersectGAN consists of two discriminators $D_1$ and $D_2$ to distinguish between generated and real samples of different domains, and three generators where the intersection generator is trained against both discriminators. And an overall adversarial loss function is defined over three generators. As a result, our proposed IntersectGAN can be trained on multiple domains of which each presents one specific attribute, and eventually eliminates the need of real sample images simultaneously possessing multiple attributes. By using the CelebFaces Attributes dataset, our proposed IntersectGAN is able to produce high quality face images possessing multiple attributes (e.g., a face with black hair and a smiling expression). Both qualitative and quantitative evaluations are conducted to compare our proposed IntersectGAN with other baseline methods. Besides, several different applications of IntersectGAN have been explored with promising results.

cs.CV↗
arXiv2018

Combining Recurrent Neural Networks and Adversarial Training for Human Motion Synthesis and Control

This paper introduces a new generative deep learning network for human motion synthesis and control. Our key idea is to combine recurrent neural networks (RNNs) and adversarial training for human motion modeling. We first describe an efficient method for training a RNNs model from prerecorded motion data. We implement recurrent neural networks with long short-term memory (LSTM) cells because they are capable of handling nonlinear dynamics and long term temporal dependencies present in human motions. Next, we train a refiner network using an adversarial loss, similar to Generative Adversarial Networks (GANs), such that the refined motion sequences are indistinguishable from real motion capture data using a discriminative network. We embed contact information into the generative deep learning model to further improve the performance of our generative model. The resulting model is appealing to motion synthesis and control because it is compact, contact-aware, and can generate an infinite number of naturally looking motions with infinite lengths. Our experiments show that motions generated by our deep learning model are always highly realistic and comparable to high-quality motion capture data. We demonstrate the power and effectiveness of our models by exploring a variety of applications, ranging from random motion synthesis, online/offline motion control, and motion filtering. We show the superiority of our generative model by comparison against baseline models.

cs.GR↗
arXiv2018

Crystalline C60 fulleride with metal inside

Endohedral metallofullerenes have been extensively studied, since the first experimental observation of La@C60 in a laser-vaporized supersonic beam in 1985. However, all of these studies have been carried out on metallofullerenes larger than C60 such as (metal)@C82, and there are no reported purified C60-based metallofullerenes except for [Li@C60]+(SbCl6)- salt. Pure (metal)@C60 has not been obtained because of their extremely high chemical reactivity. We report here the first isolation, structural determination and electromagnetic properties of crystalline C60-based metallofullerenes, Gd@C60(CF3)5 and La@C60(CF3)5. Synchrotron X-ray single-crystal diffraction reveals that La and Gd atoms are indeed encapsulated in the Ih-C60 fullerene. The HOMO-LUMO gaps of Gd@C60 and La@C60 are significantly widened by an order of magnitude with addition of CF3- groups. Magnetic measurements show the presense of a weak antiferromagnetic coupling in Gd@C60(CF3)3 crystals at low temperatures. Gd@C60 and La@C60 might exhibit superconductivity as the electronic structures resemble those of superconducting alkali-doped C60 fullerides.

cond-mat.mtrl-sci↗
arXiv2017

Approaching quantum anomalous Hall effect in proximity-coupled YIG/graphene/h-BN sandwich structure

Quantum anomalous Hall state is expected to emerge in Dirac electron systems such as graphene under both sufficiently strong exchange and spin-orbit interactions. In pristine graphene, neither interaction exists; however, both interactions can be acquired by coupling graphene to a magnetic insulator (MI) as revealed by the anomalous Hall effect. Here, we show enhanced magnetic proximity coupling by sandwiching graphene between a ferrimagnetic insulator yttrium iron garnet (YIG) and hexagonal-boron nitride (h-BN) which also serves as a top gate dielectric. By sweeping the top-gate voltage, we observe Fermi level-dependent anomalous Hall conductance. As the Dirac point is approached from both electron and hole sides, the anomalous Hall conductance reaches 1/4 of the quantum anomalous Hall conductance 2e2/h. The exchange coupling strength is determined to be as high as 27 meV from the transition temperature of the induced magnetic phase. YIG/graphene/h-BN is an excellent heterostructure for demonstrating proximity-induced interactions in two-dimensional electron systems.

cond-mat.mes-hall↗
arXiv2017

Filtering Signal Processes in Molecular Multimedia Memristors

To obtain precisely controllable, robust as well as reproduceable memristor for efficient neuromorphic computing still very challenging. Molecular tailoring aims at obtaining the much more flexibly tuning plasticity has recently generated significant interest as new paradigms toward the realization of novel memristor-based synapses. Herein, inspired by the deliberate oxygen transport carried by the hemoglobin in our blood circulation, we report a novel molecular-regulated electronic/ionic semiconducting (MEIS) platform ITO/MTPP/Al2O3-x/Al with a series of metallophorphyrins (MTPPs) to delicately regulate the ionic migration for robust molecular multimedia memristors. The stable pinched hysteresis resulted from the coordination-regulated ionic migration was verified by different device structures, operation modes, as well as the characterizations of scanning transmission electron microscopy with energy-dispersive X-ray spectroscopy (STEM-EDX) and X-ray photoelectron spectroscopy (XPS). Metal coordination-dependent device parameters such as potential and depression as well as retention curves further support the correlation between the coordination and stimulating flux-dependent memristive behaviors. In the 5,10,15,20-tetraphenyl-21H,23H-porphyrin zinc(II) (ZnTPP) synapse, we implement versatile emulations, mainly including transition from short-term memory (STM) to long-term-memory (LTM), learning experience and activity-dependent learning-memory process in integrated neuromorphic configurations based on the biological Hebbian rules, and develop the fresh Spike-Amplitude-Dependent-Plasticity (SADP) with the applications of signal filtering and habituation and sensitization which are beyond the prevalent Hebbian rules.

cond-mat.mes-hall↗
arXiv2016

Knowledge-based machine learning methods for macromolecular 3D structure prediction

Predicting the 3D structure of a macromolecule, such as a protein or an RNA molecule, is ranked top among the most difficult and attractive problems in bioinformatics and computational biology. Its importance comes from the relationship between the 3D structure and the function of a given protein or RNA. 3D structures also help to find the ligands of the protein, which are usually small molecules, a key step in drug design. Unfortunately, there is no shortcut to accurately obtain the 3D structure of a macromolecule. Many physical measurements of macromolecular 3D structures cannot scale up, due to their large labor costs and the requirements for lab conditions. In recent years, computational methods have made huge progress due to advance in computation speed and machine learning methods. These methods only need the sequence information to predict 3D structures by employing various mathematical models and machine learning methods. The success of computational methods is highly dependent on a large database of the proteins and RNA with known structures. However, the performance of computational methods are always expected to be improved. There are several reasons for this. First, we are facing, and will continue to face sparseness of data.Secondly, the 3D structure space is too large for our computational capability. The two obstacles can be removed by knowledge-based methods, which combine knowledge learned from the known structures and biologists' knowledge of the folding process of protein or RNA. In the dissertation, I will present my results in building a knowledge-based method by using machine learning methods to tackle this problem. My methods include the knowledge constraints on intermediate states, which can highly reduce the solution space of a protein or RNA, in turn increasing the efficiency of the structure folding method and improving its accuracy.

q-bio.BM↗
arXiv2016

Matrix Neural Networks

Traditional neural networks assume vectorial inputs as the network is arranged as layers of single line of computing units called neurons. This special structure requires the non-vectorial inputs such as matrices to be converted into vectors. This process can be problematic. Firstly, the spatial information among elements of the data may be lost during vectorisation. Secondly, the solution space becomes very large which demands very special treatments to the network parameters and high computational cost. To address these issues, we propose matrix neural networks (MatNet), which takes matrices directly as inputs. Each neuron senses summarised information through bilinear mapping from lower layer units in exactly the same way as the classic feed forward neural networks. Under this structure, back prorogation and gradient descent combination can be utilised to obtain network parameters efficiently. Furthermore, it can be conveniently extended for multimodal inputs. We apply MatNet to MNIST handwritten digits classification and image super resolution tasks to show its effectiveness. Without too much tweaking MatNet achieves comparable performance as the state-of-the-art methods in both tasks with considerably reduced complexity.

cs.LG↗
arXiv2015

Protein Contact Prediction by Integrating Joint Evolutionary Coupling Analysis and Supervised Learning

Protein contacts contain important information for protein structure and functional study, but contact prediction from sequence remains very challenging. Both evolutionary coupling (EC) analysis and supervised machine learning methods are developed to predict contacts, making use of different types of information, respectively. This paper presents a group graphical lasso (GGL) method for contact prediction that integrates joint multi-family EC analysis and supervised learning. Different from existing single-family EC analysis that uses residue co-evolution information in only the target protein family, our joint EC analysis uses residue co-evolution in both the target family and its related families, which may have divergent sequences but similar folds. To implement joint EC analysis, we model a set of related protein families using Gaussian graphical models (GGM) and then co-estimate their precision matrices by maximum-likelihood, subject to the constraint that the precision matrices shall share similar residue co-evolution patterns. To further improve the accuracy of the estimated precision matrices, we employ a supervised learning method to predict contact probability from a variety of evolutionary and non-evolutionary information and then incorporate the predicted probability as prior into our GGL framework. Experiments show that our method can predict contacts much more accurately than existing methods, and that our method performs better on both conserved and family-specific contacts.

q-bio.QM↗
arXiv2014

Proximity-induced ferromagnetism in graphene revealed by anomalous Hall effect

We demonstrate the anomalous Hall effect (AHE) in single-layer graphene exchange-coupled to an atomically flat yttrium iron garnet (YIG) ferromagnetic thin film. The anomalous Hall conductance has magnitude of ~0.09(2e2/h) at low temperatures and is measurable up to ~ 300 K. Our observations indicate not only proximity-induced ferromagnetism in graphene/YIG with large exchange interaction, but also enhanced spin-orbit coupling which is believed to be inherently weak in ideal graphene. The proximity-induced ferromagnetic order in graphene can lead to novel transport phenomena such as the quantized AHE which are potentially useful for spintronics.

cond-mat.mes-hall↗
arXiv2014

MRFalign: Protein Homology Detection through Alignment of Markov Random Fields

Sequence-based protein homology detection has been extensively studied and so far the most sensitive method is based upon comparison of protein sequence profiles, which are derived from multiple sequence alignment (MSA) of sequence homologs in a protein family. A sequence profile is usually represented as a position-specific scoring matrix (PSSM) or an HMM (Hidden Markov Model) and accordingly PSSM-PSSM or HMM-HMM comparison is used for homolog detection. This paper presents a new homology detection method MRFalign, consisting of three key components: 1) a Markov Random Fields (MRF) representation of a protein family; 2) a scoring function measuring similarity of two MRFs; and 3) an efficient ADMM (Alternating Direction Method of Multipliers) algorithm aligning two MRFs. Compared to HMM that can only model very short-range residue correlation, MRFs can model long-range residue interaction pattern and thus, encode information for the global 3D structure of a protein family. Consequently, MRF-MRF comparison for remote homology detection shall be much more sensitive than HMM-HMM or PSSM-PSSM comparison. Experiments confirm that MRFalign outperforms several popular HMM or PSSM-based methods in terms of both alignment accuracy and remote homology detection and that MRFalign works particularly well for mainly beta proteins. For example, tested on the benchmark SCOP40 (8353 proteins) for homology detection, PSSM-PSSM and HMM-HMM succeed on 48% and 52% of proteins, respectively, at superfamily level, and on 15% and 27% of proteins, respectively, at fold level. In contrast, MRFalign succeeds on 57.3% and 42.5% of proteins at superfamily and fold level, respectively. This study implies that long-range residue interaction patterns are very helpful for sequence-based homology detection. The software is available for download at http://raptorx.uchicago.edu/download/.

q-bio.QM↗
arXiv2013

Predicting protein contact map using evolutionary and physical constraints by integer programming (extended version)

Motivation. Protein contact map describes the pairwise spatial and functional relationship of residues in a protein and contains key information for protein 3D structure prediction. Although studied extensively, it remains very challenging to predict contact map using only sequence information. Most existing methods predict the contact map matrix element-by-element, ignoring correlation among contacts and physical feasibility of the whole contact map. A couple of recent methods predict contact map based upon residue co-evolution, taking into consideration contact correlation and enforcing a sparsity restraint, but these methods require a very large number of sequence homologs for the protein under consideration and the resultant contact map may be still physically unfavorable. Results. This paper presents a novel method PhyCMAP for contact map prediction, integrating both evolutionary and physical restraints by machine learning and integer linear programming (ILP). The evolutionary restraints include sequence profile, residue co-evolution and context-specific statistical potential. The physical restraints specify more concrete relationship among contacts than the sparsity restraint. As such, our method greatly reduces the solution space of the contact map matrix and thus, significantly improves prediction accuracy. Experimental results confirm that PhyCMAP outperforms currently popular methods no matter how many sequence homologs are available for the protein under consideration. PhyCMAP can predict contacts within minutes after PSIBLAST search for sequence homologs is done, much faster than the two recent methods PSICOV and EvFold. See http://raptorx.uchicago.edu for the web server.

q-bio.QM↗