arXiv subjects

Bing Yang

Publications and source records attributed to Bing Yang.

At least 55 records · Page 3Linked to original sources

arXiv2023

xTrimoABFold: De novo Antibody Structure Prediction without MSA

In the field of antibody engineering, an essential task is to design a novel antibody whose paratopes bind to a specific antigen with correct epitopes. Understanding antibody structure and its paratope can facilitate a mechanistic understanding of its function. Therefore, antibody structure prediction from its sequence alone has always been a highly valuable problem for de novo antibody design. AlphaFold2, a breakthrough in the field of structural biology, provides a solution to predict protein structure based on protein sequences and computationally expensive coevolutionary multiple sequence alignments (MSAs). However, the computational efficiency and undesirable prediction accuracy of antibodies, especially on the complementarity-determining regions (CDRs) of antibodies limit their applications in the industrially high-throughput drug design. To learn an informative representation of antibodies, we employed a deep antibody language model (ALM) on curated sequences from the observed antibody space database via a transformer model. We also developed a novel model named xTrimoABFold to predict antibody structure from antibody sequence based on the pretrained ALM as well as efficient evoformers and structural modules. The model was trained end-to-end on the antibody structures in PDB by minimizing the ensemble loss of domain-specific focal loss on CDR and the frame-aligned point loss. xTrimoABFold outperforms AlphaFold2 and other protein language model based SOTAs, e.g., OmegaFold, HelixFold-Single, and IgFold with a large significant margin (30+\% improvement on RMSD) while performing 151 times faster than AlphaFold2. To the best of our knowledge, xTrimoABFold achieved state-of-the-art antibody structure prediction. Its improvement in both accuracy and efficiency makes it a valuable tool for de novo antibody design and could make further improvements in immuno-theory.

q-bio.QM↗

arXiv2023

High-order adaptive multiresolution wavelet upwind schemes for hyperbolic conservation laws

A system of high-order adaptive multiresolution wavelet collocation upwind schemes are developed for the solution of hyperbolic conservation laws. A couple of asymmetrical wavelet bases with interpolation property are built to realize the upwind property, and address the nonlinearity in the hyperbolic problems. An adaptive algorithm based on multiresolution analysis in wavelet theory is designed to capture moving shock waves and distinguish new localized steep regions. An integration average reconstruction method is proposed based on the Lebesgue differentiation theorem to suppress the Gibbs phenomenon. All these numerical techniques enable the wavelet collocation upwind scheme to provide a general framework for devising satisfactory adaptive wavelet upwind methods with high-order accuracy. Several benchmark tests for 1D hyperbolic problems are carried out to verify the accuracy and efficiency of the present wavelet schemes.

math.NA↗

arXiv2022

Large-Scale $2+1$D $\mathrm{U}(1)$ Gauge Theory with Dynamical Matter in a Cold-Atom Quantum Simulator

A major driver of quantum-simulator technology is the prospect of probing high-energy phenomena in synthetic quantum matter setups at a high level of control and tunability. Here, we propose an experimentally feasible realization of a large-scale $2+1$D $\mathrm{U}(1)$ gauge theory with dynamical matter and gauge fields in a cold-atom quantum simulator with spinless bosons. We present the full mapping of the corresponding Gauss's law onto the bosonic computational basis. We then show that the target gauge theory can be faithfully realized and stabilized by an emergent gauge protection term in a two-dimensional single-species Bose--Hubbard optical Lieb superlattice with two spatial periods along either direction, thereby requiring only moderate experimental resources already available in current cold-atom setups. Using infinite matrix product states, we calculate numerical benchmarks for adiabatic sweeps and global quench dynamics that further confirm the fidelity of the mapping. Our work brings quantum simulators of gauge theories a significant step forward in terms of investigating particle physics in higher spatial dimensions, and is readily implementable in existing cold-atom platforms.

cond-mat.quant-gas↗

arXiv2022

AdaVITS: Tiny VITS for Low Computing Resource Speaker Adaptation

Speaker adaptation in text-to-speech synthesis (TTS) is to finetune a pre-trained TTS model to adapt to new target speakers with limited data. While much effort has been conducted towards this task, seldom work has been performed for low computational resource scenarios due to the challenges raised by the requirement of the lightweight model and less computational complexity. In this paper, a tiny VITS-based TTS model, named AdaVITS, for low computing resource speaker adaptation is proposed. To effectively reduce parameters and computational complexity of VITS, an iSTFT-based wave construction decoder is proposed to replace the upsampling-based decoder which is resource-consuming in the original VITS. Besides, NanoFlow is introduced to share the density estimate across flow blocks to reduce the parameters of the prior encoder. Furthermore, to reduce the computational complexity of the textual encoder, scaled-dot attention is replaced with linear attention. To deal with the instability caused by the simplified model, instead of using the original text encoder, phonetic posteriorgram (PPG) is utilized as linguistic feature via a text-to-PPG module, which is then used as input for the encoder. Experiment shows that AdaVITS can generate stable and natural speech in speaker adaptation with 8.97M model parameters and 0.72GFlops computational complexity.

cs.SD↗

arXiv2022

Thermalization dynamics of a gauge theory on a quantum simulator

Gauge theories form the foundation of modern physics, with applications ranging from elementary particle physics and early-universe cosmology to condensed matter systems. We perform quantum simulations of the unitary dynamics of a U(1) symmetric gauge field theory and demonstrate emergent irreversible behavior. The highly constrained gauge theory dynamics is encoded in a one-dimensional Bose--Hubbard simulator, which couples fermionic matter fields through dynamical gauge fields. We investigate global quantum quenches and the equilibration to a steady state well approximated by a thermal ensemble. Our work may enable the investigation of elusive phenomena, such as Schwinger pair production and string-breaking, and paves the way for simulating more complex higher-dimensional gauge theories on quantum synthetic matter devices.

cond-mat.quant-gas↗

arXiv2022

Multi-Scale Spatial Temporal Graph Convolutional Network for Skeleton-Based Action Recognition

Graph convolutional networks have been widely used for skeleton-based action recognition due to their excellent modeling ability of non-Euclidean data. As the graph convolution is a local operation, it can only utilize the short-range joint dependencies and short-term trajectory but fails to directly model the distant joints relations and long-range temporal information that are vital to distinguishing various actions. To solve this problem, we present a multi-scale spatial graph convolution (MS-GC) module and a multi-scale temporal graph convolution (MT-GC) module to enrich the receptive field of the model in spatial and temporal dimensions. Concretely, the MS-GC and MT-GC modules decompose the corresponding local graph convolution into a set of sub-graph convolution, forming a hierarchical residual architecture. Without introducing additional parameters, the features will be processed with a series of sub-graph convolutions, and each node could complete multiple spatial and temporal aggregations with its neighborhoods. The final equivalent receptive field is accordingly enlarged, which is capable of capturing both short- and long-range dependencies in spatial and temporal domains. By coupling these two modules as a basic block, we further propose a multi-scale spatial temporal graph convolutional network (MST-GCN), which stacks multiple blocks to learn effective motion representations for action recognition. The proposed MST-GCN achieves remarkable performance on three challenging benchmark datasets, NTU RGB+D, NTU-120 RGB+D and Kinetics-Skeleton, for skeleton-based action recognition.

cs.CV↗

arXiv2022

Tuning the Topological $θ$-Angle in Cold-Atom Quantum Simulators of Gauge Theories

The topological $θ$-angle in gauge theories engenders a series of fundamental phenomena, including violations of charge-parity (CP) symmetry, dynamical topological transitions, and confinement--deconfinement transitions. At the same time, it poses major challenges for theoretical studies, as it implies a sign problem in numerical simulations. Analog quantum simulators open the promising prospect of treating quantum many-body systems with such topological terms, but, contrary to their digital counterparts, they have not yet demonstrated the capacity to control the $θ$-angle. Here, we demonstrate how a tunable topological $θ$-term can be added to a prototype theory with $\mathrm{U}(1)$ gauge symmetry, a discretized version of quantum electrodynamics in one spatial dimension. As we show, the model can be realized experimentally in a single-species Bose--Hubbard model in an optical superlattice with three different spatial periods, thus requiring only standard experimental resources. Through numerical calculations obtained from the time-dependent density matrix renormalization group method and exact diagonalization, we benchmark the model system, and illustrate how salient effects due to the $θ$-term can be observed. These include charge confinement, the weakening of quantum many-body scarring, as well as the disappearance of Coleman's phase transition due to explicit breaking of CP symmetry. This work opens the door towards studying the rich physics of topological gauge-theory terms in large-scale cold-atom quantum simulators.

cond-mat.quant-gas↗

arXiv2022

Non-neglectable entropy effect on sintering of supported nanoparticles

Sintering refers to particle coalescence by heat, which has been known as a thermal phenomenon involving all aspects of natural science for centuries. It is particularly important in heterogeneous catalysis because normally sintering results in deactivation of the catalysts. In previous studies, the enthalpy contribution was considered to be dominant in sintering and the entropy effect is generally considered neglectable. However, we unambiguously demonstrate in this work that entropy could prevail over the enthalpy contribution to dominate the sintering behavior of supported nanoparticles (NPs) by designed experiments and improved theoretical framework. Using in situ Cs-corrected environmental scanning transmission electron microscopy and synchrotron-based ambient pressure X-ray photoelectron spectroscopy, we observe the unprecedent entropy-driven phenomenon that supported NPs reversibly redisperse upon heating and sinter upon cooling in three systems (Pd-CeO2, Cu-TiO2, Ag-TiO2). We quantitatively show that the configurational entropy of highly dispersed ad-atoms is large enough to reverse their sintering tendency at the elevated temperature. This work reshapes the basic understanding of sintering at the nanoscale and opens the door for various de-novo designs of thermodynamically stable nanocatalysts.

cond-mat.mtrl-sci↗

arXiv2022

Learning Deep Direct-Path Relative Transfer Function for Binaural Sound Source Localization

Direct-path relative transfer function (DP-RTF) refers to the ratio between the direct-path acoustic transfer functions of two microphone channels. Though DP-RTF fully encodes the sound spatial cues and serves as a reliable localization feature, it is often erroneously estimated in the presence of noise and reverberation. This paper proposes to learn DP-RTF with deep neural networks for robust binaural sound source localization. A DP-RTF learning network is designed to regress the binaural sensor signals to a real-valued representation of DP-RTF. It consists of a branched convolutional neural network module to separately extract the inter-channel magnitude and phase patterns, and a convolutional recurrent neural network module for joint feature learning. To better explore the speech spectra to aid the DP-RTF estimation, a monaural speech enhancement network is used to recover the direct-path spectrograms from the noisy ones. The enhanced spectrograms are stacked onto the noisy spectrograms to act as the input of the DP-RTF learning network. We train one unique DP-RTF learning network using many different binaural arrays to enable the generalization of DP-RTF learning across arrays. This way avoids time-consuming training data collection and network retraining for a new array, which is very useful in practical application. Experimental results on both simulated and real-world data show the effectiveness of the proposed method for direction of arrival (DOA) estimation in the noisy and reverberant environment, and a good generalization ability to unseen binaural arrays.

cs.SD↗

arXiv2022

SRP-DNN: Learning Direct-Path Phase Difference for Multiple Moving Sound Source Localization

Multiple moving sound source localization in real-world scenarios remains a challenging issue due to interaction between sources, time-varying trajectories, distorted spatial cues, etc. In this work, we propose to use deep learning techniques to learn competing and time-varying direct-path phase differences for localizing multiple moving sound sources. A causal convolutional recurrent neural network is designed to extract the direct-path phase difference sequence from signals of each microphone pair. To avoid the assignment ambiguity and the problem of uncertain output-dimension encountered when simultaneously predicting multiple targets, the learning target is designed in a weighted sum format, which encodes source activity in the weight and direct-path phase differences in the summed value. The learned direct-path phase differences for all microphone pairs can be directly used to construct the spatial spectrum according to the formulation of steered response power (SRP). This deep neural network (DNN) based SRP method is referred to as SRP-DNN. The locations of sources are estimated by iteratively detecting and removing the dominant source from the spatial spectrum, in which way the interaction between sources is reduced. Experimental results on both simulated and real-world data show the superiority of the proposed method in the presence of noise and reverberation.

cs.SD↗

arXiv2021

AlCrO protected textured stainless steel surface for high temperature solar selective absorber applications

The diffusion of substrate material into absorbing layer and oxidation of metal substrate or cermet metal nanoparticles at high temperatures are known as inevitable problems of the solar selective absorbers. In this study, we consider the use of textured stainless steel (SS) surface coated with a protective AlCr oxide layer as a high temperature solar selective absorber. The textured SS surface was prepared by ion etching techniques and AlCr oxide protective layer was deposited by RF magnetron sputtering. The absorptivity and emissivity of the as-prepared absorbers were 0.86-0.92 and 0.151-0.168, respectively. In order to evaluate the thermal stability, the absorbers were annealed at 600-800 C for different time in ambient atmosphere. Absorbers demonstrated a red shift of the onset of the reflectivity at all annealing temperatures. The high activation energy of 315 kJ/mol was calculated. The service lifetime of the absorbers at 500 C was estimated to be about 100 years and at 700 and 800 C the absorbers were stable about 50 and 1 hours, respectively. A detailed examination of the annealed absorber surface revealed growth of surface Mn3O4 nanocrystals, which resulted in observed change of the reflectance spectra, while the textured surface morphology had no significant change. The results show that the protective textured surface has much higher thermal stability in air than iron based cermet absorbers.

physics.chem-ph↗

arXiv2021

Realization of a Bosonic Antiferromagnet

Quantum antiferromagnets are of broad interest in condensed matter physics as they provide a platform for studying exotic many-body states including spin liquids and high-temperature superconductors. Here, we report on the creation of a one-dimensional Heisenberg antiferromagnet with ultracold bosons. In a two-component Bose-Hubbard system, we switch the sign of the spin-exchange interaction and realize the isotropic antiferromagnetic Heisenberg model in an extended 70-site chain. Starting from a low-entropy Néel-ordered state, we use optimized adiabatic passage to approach the bosonic antiferromagnet. We demonstrate the establishment of antiferromagnetism by probing the evolution of the staggered magnetization and spin correlations of the system. Compared with condensed matter systems, ultracold gases in optical lattices can be microscopically engineered and measured, offering significant advantages for exploring bosonic magnetism and spin dynamics.

cond-mat.quant-gas↗

arXiv2020

Observation of gauge invariance in a 71-site Bose-Hubbard quantum simulator

The modern description of elementary particles, as formulated in the Standard Model of particle physics, is built on gauge theories. Gauge theories implement fundamental laws of physics by local symmetry constraints. For example, in quantum electrodynamics, Gauss's law introduces an intrinsic local relation between charged matter and electromagnetic fields, which protects many salient physical properties including massless photons and a long-ranged Coulomb law. Solving gauge theories by classical computers is an extremely arduous task, which has stimulated a vigorous effort to simulate gauge-theory dynamics in microscopically engineered quantum devices. Previous achievements implemented density-dependent Peierls phases without defining a local symmetry, realized mappings onto effective models to integrate out either matter or electric fields, or were limited to very small systems. The essential gauge symmetry has not been observed experimentally. Here, we report the quantum simulation of an extended U(1) lattice gauge theory, and experimentally quantify the gauge invariance in a many-body system comprising matter and gauge fields. These are realized in defect-free arrays of bosonic atoms in an optical superlattice of 71 sites. We demonstrate full tunability of the model parameters and benchmark the matter--gauge interactions by sweeping across a quantum phase transition. Enabled by high-fidelity manipulation techniques, we measure the degree to which Gauss's law is violated by extracting probabilities of locally gauge-invariant states from correlated atom occupations. Our work provides a way to explore gauge symmetry in the interplay of fundamental particles using controllable large-scale quantum simulators.

cond-mat.quant-gas↗

arXiv2020

Cooling and entangling ultracold atoms in optical lattices

Scalable, coherent many-body systems can enable the realization of previously unexplored quantum phases and have the potential to exponentially speed up information processing. Thermal fluctuations are negligible and quantum effects govern the behavior of such systems with extremely low temperature. We report the cooling of a quantum simulator with 10,000 atoms and mass production of high-fidelity entangled pairs. In a two-dimensional plane, we cool Mott insulator samples by immersing them into removable superfluid reservoirs, achieving an entropy per particle of $1.9^{+1.7}_{-0.4} \times 10^{-3} k_{\text{B}}$. The atoms are then rearranged into a two-dimensional lattice free of defects. We further demonstrate a two-qubit gate with a fidelity of 0.993 $\pm$ 0.001 for entangling 1250 atom pairs. Our results offer a setting for exploring low-energy many-body phases and may enable the creation of large-scale entanglement

cond-mat.quant-gas↗

arXiv2020

Robustness of gauge-invariant dynamics against defects in ultracold-atom gauge theories

Recent years have seen strong progress in quantum simulation of gauge-theory dynamics using ultracold-atom experiments. A principal challenge in these efforts is the certification of gauge invariance, which has recently been realized in [B.~Yang et al., arXiv:2003.08945]. One major but poorly investigated experimental source of gauge-invariance violation is an imperfect preparation of the initial state. Using the time-dependent density-matrix renormalization group, we analyze the robustness of gauge-invariant dynamics against potential preparation defects in the above ultracold-atom implementation of a $\mathrm{U}(1)$ gauge theory. We find defects related to an erroneous initialization of matter fields to be innocuous, as the associated gauge-invariance violation remains strongly localized throughout the time evolution. A defect due to faulty initialization of the gauge field leads to a mild proliferation of the associated violation. Furthermore, we characterize the influence of immobile and mobile defects by monitoring the spread of entanglement entropy. Overall, our results indicate that the aforementioned experimental realization exhibits a high level of fidelity in the gauge invariance of its dynamics at all evolution times. Our work provides strong evidence that ultracold-atom setups can serve as an extremely reliable framework for the quantum simulation of gauge-theory dynamics.

cond-mat.quant-gas↗

arXiv2019

SkrGAN: Sketching-rendering Unconditional Generative Adversarial Networks for Medical Image Synthesis

Generative Adversarial Networks (GANs) have the capability of synthesizing images, which have been successfully applied to medical image synthesis tasks. However, most of existing methods merely consider the global contextual information and ignore the fine foreground structures, e.g., vessel, skeleton, which may contain diagnostic indicators for medical image analysis. Inspired by human painting procedure, which is composed of stroking and color rendering steps, we propose a Sketching-rendering Unconditional Generative Adversarial Network (SkrGAN) to introduce a sketch prior constraint to guide the medical image generation. In our SkrGAN, a sketch guidance module is utilized to generate a high quality structural sketch from random noise, then a color render mapping is used to embed the sketch-based representations and resemble the background appearances. Experimental results show that the proposed SkrGAN achieves the state-of-the-art results in synthesizing images for various image modalities, including retinal color fundus, X-Ray, Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). In addition, we also show that the performances of medical image segmentation method have been improved by using our synthesized images as data augmentation.

cs.CV↗

arXiv2018

Battery-like Supercapacitors from Vertically Aligned Carbon Nanofibers Coated Diamond: Design and Demonstrator

Battery-like supercapacitors feature high power and energy densities as well as long-term capacitance retention. The utilized capacitor electrodes are thus better to have large surface areas, high conductivity, high stability, and importantly be of binder free. Herein, vertically aligned carbon nanofibers (CNFs) coated boron-doped diamonds (BDD) are employed as the capacitor electrodes to construct battery-like supercapacitors. Grown via a thermal chemical vapor deposition technique, these CNFs/BDD hybrid films are binder free and own porous structures, resulting in large surface areas. Meanwhile, the containment of graphene layers and copper metal catalysts inside CNFs/BDD leads to their high conductivity. Electric double layer capacitors (EDLCs) and pseudocapacitors (PCs) are then constructed in the inert electrolyte (1.0 M H2SO4 solution) and in the redox-active electrolyte (1.0 M Na2SO4 + 0.05 M Fe(CN)63-/4-), respectively. For assembled two-electrode symmetrical supercapacitor devices, the capacitances of EDLC and PC devices reach 30 and 48 mF cm-2 at 10 mV s-1, respectively. They remain constant even after 10 000 cycles. The power densities are 27.3 kW kg-1 and 25.3 kW kg-1 for EDLC and PC devices, together with their energy densities of 22.9 Wh kg-1 and 44.1 Wh kg-1, respectively. The performance of formed EDLC and PC devices is comparable to market-available batteries. Therefore, the vertically aligned CNFs/BDD hybrid film is a suitable capacitor electrode material to construct high-performance battery-like and industry-orientated supercapacitors for flexible power devices.

physics.app-ph↗

arXiv2017

Engineering topological edge states in two dimensional magnetic photonic crystal

Based on a perturbative approach, we propose a simple and efficient method to engineer topological edge states in two dimensional magnetic photonic crystals. The topological edge states in the microstructures can be constructed and varied by altering the parameters of the microstructure according to the fieldenergy distributions of the Bloch states at the related Bloch wave vectors. The validity of the proposed method has been demonstrated by exact numerical calculations through three concrete examples. Our method makes the topological edge states 'designable'.

physics.optics↗