arXiv subjects

Hui Li

Publications and source records attributed to Hui Li.

At least 595 records · Page 33Linked to original sources

arXiv2021

Magnetic Energy Release, Plasma Dynamics, and Particle Acceleration during Relativistic Turbulent Magnetic Reconnection

In strongly magnetized astrophysical plasma systems, magnetic reconnection is believed to be a primary process during which explosive energy release and particle acceleration occur, leading to significant high-energy emission. Past years have witnessed active development of kinetic modeling of relativistic magnetic reconnection, supporting this magnetically dominated scenario. A much less explored issue is the consequence of 3D dynamics, where turbulent structures are naturally generated as various types of instabilities develop. This paper presents a series of 3D, fully-kinetic simulations of relativistic turbulent magnetic reconnection (RTMR) in positron-electron plasmas with system domains much larger than kinetic scales. Our simulations start from a force-free current sheet with several different modes of long wavelength magnetic field perturbations, which drive additional turbulence in the reconnection region. Because of this, the current layer breaks up and the reconnection region quickly evolves into a turbulent layer filled with coherent structures such as flux ropes and current sheets. We find that plasma dynamics in RTMR is vastly different from their 2D counterparts in many aspects. The flux ropes evolve rapidly after their generation, and can be completely disrupted due to the secondary kink instability. This turbulent evolution leads to superdiffusion behavior of magnetic field lines as seen in MHD studies of turbulent reconnection. Meanwhile, nonthermal particle acceleration and energy-release time scale can be very fast and do not strongly depend on the turbulence amplitude. The main acceleration mechanism is a Fermi-like acceleration process supported by the motional electric field, whereas the non-ideal electric field acceleration plays a subdominant role. We discuss possible observational implications of 3D RTMR in high-energy astrophysics.

astro-ph.HE↗

arXiv2021

Metastability for the dissipative quasi-geostrophic equation and the non-local enhancement

In this paper, we study the metastability for the 2-D linearized dissipative quasi-geostrophic equation with small viscosity $ν$ around the quasi steady state $θ_{sin}=e^{-νt}\sin y$. We proved the linear enhanced dissipation and obtained the dissipation rate. Moreover, the new non-local enhancement phenomenon was discovered and discussed. Precisely we showed that the non-local term re-enhances the enhanced diffusion effect by the shear-diffusion mechanism.

math.AP↗

arXiv2021

Attention-Guided Progressive Neural Texture Fusion for High Dynamic Range Image Restoration

High Dynamic Range (HDR) imaging via multi-exposure fusion is an important task for most modern imaging platforms. In spite of recent developments in both hardware and algorithm innovations, challenges remain over content association ambiguities caused by saturation, motion, and various artifacts introduced during multi-exposure fusion such as ghosting, noise, and blur. In this work, we propose an Attention-guided Progressive Neural Texture Fusion (APNT-Fusion) HDR restoration model which aims to address these issues within one framework. An efficient two-stream structure is proposed which separately focuses on texture feature transfer over saturated regions and multi-exposure tonal and texture feature fusion. A neural feature transfer mechanism is proposed which establishes spatial correspondence between different exposures based on multi-scale VGG features in the masked saturated HDR domain for discriminative contextual clues over the ambiguous image areas. A progressive texture blending module is designed to blend the encoded two-stream features in a multi-scale and progressive manner. In addition, we introduce several novel attention mechanisms, i.e., the motion attention module detects and suppresses the content discrepancies among the reference images; the saturation attention module facilitates differentiating the misalignment caused by saturation from those caused by motion; and the scale attention module ensures texture blending consistency between different coder/decoder scales. We carry out comprehensive qualitative and quantitative evaluations and ablation studies, which validate that these novel modules work coherently under the same framework and outperform state-of-the-art methods.

eess.IV↗

arXiv2021

SHORING: Design Provable Conditional High-Order Interaction Network via Symbolic Testing

Deep learning provides a promising way to extract effective representations from raw data in an end-to-end fashion and has proven its effectiveness in various domains such as computer vision, natural language processing, etc. However, in domains such as content/product recommendation and risk management, where sequence of event data is the most used raw data form and experts derived features are more commonly used, deep learning models struggle to dominate the game. In this paper, we propose a symbolic testing framework that helps to answer the question of what kinds of expert-derived features could be learned by a neural network. Inspired by this testing framework, we introduce an efficient architecture named SHORING, which contains two components: \textit{event network} and \textit{sequence network}. The \textit{event} network learns arbitrarily yet efficiently high-order \textit{event-level} embeddings via a provable reparameterization trick, the \textit{sequence} network aggregates from sequence of \textit{event-level} embeddings. We argue that SHORING is capable of learning certain standard symbolic expressions which the standard multi-head self-attention network fails to learn, and conduct comprehensive experiments and ablation studies on four synthetic datasets and three real-world datasets. The results show that SHORING empirically outperforms the state-of-the-art methods.

cs.LG↗

arXiv2021

Towards End-to-End Text Spotting in Natural Scenes

Text spotting in natural scene images is of great importance for many image understanding tasks. It includes two sub-tasks: text detection and recognition. In this work, we propose a unified network that simultaneously localizes and recognizes text with a single forward pass, avoiding intermediate processes such as image cropping and feature re-calculation, word separation, and character grouping. In contrast to existing approaches that consider text detection and recognition as two distinct tasks and tackle them one by one, the proposed framework settles these two tasks concurrently. The whole framework can be trained end-to-end and is able to handle text of arbitrary shapes. The convolutional features are calculated only once and shared by both detection and recognition modules. Through multi-task training, the learned features become more discriminate and improve the overall performance. By employing the $2$D attention model in word recognition, the irregularity of text can be robustly addressed. It provides the spatial location for each character, which not only helps local feature extraction in word recognition, but also indicates an orientation angle to refine text localization. Our proposed method has achieved state-of-the-art performance on several standard text spotting benchmarks, including both regular and irregular ones.

cs.CV↗

arXiv2021

MatchVIE: Exploiting Match Relevancy between Entities for Visual Information Extraction

Visual Information Extraction (VIE) task aims to extract key information from multifarious document images (e.g., invoices and purchase receipts). Most previous methods treat the VIE task simply as a sequence labeling problem or classification problem, which requires models to carefully identify each kind of semantics by introducing multimodal features, such as font, color, layout. But simply introducing multimodal features couldn't work well when faced with numeric semantic categories or some ambiguous texts. To address this issue, in this paper we propose a novel key-value matching model based on a graph neural network for VIE (MatchVIE). Through key-value matching based on relevancy evaluation, the proposed MatchVIE can bypass the recognitions to various semantics, and simply focuses on the strong relevancy between entities. Besides, we introduce a simple but effective operation, Num2Vec, to tackle the instability of encoded values, which helps model converge more smoothly. Comprehensive experiments demonstrate that the proposed MatchVIE can significantly outperform previous methods. Notably, to the best of our knowledge, MatchVIE may be the first attempt to tackle the VIE task by modeling the relevancy between keys and values and it is a good complement to the existing methods.

cs.CV↗

arXiv2021

Sequential Recommendation in Online Games with Multiple Sequences, Tasks and User Levels

Online gaming is growing faster than ever before, with increasing challenges of providing better user experience. Recommender systems (RS) for online games face unique challenges since they must fulfill players' distinct desires, at different user levels, based on their action sequences of various action types. Although many sequential RS already exist, they are mainly single-sequence, single-task, and single-user-level. In this paper, we introduce a new sequential recommendation model for multiple sequences, multiple tasks, and multiple user levels (abbreviated as M$^3$Rec) in Tencent Games platform, which can fully utilize complex data in online games. We leverage Graph Neural Network and multi-task learning to design M$^3$Rec in order to model the complex information in the heterogeneous sequential recommendation scenario of Tencent Games. We verify the effectiveness of M$^3$Rec on three online games of Tencent Games platform, in both offline and online evaluations. The results show that M$^3$Rec successfully addresses the challenges of recommendation in online games, and it generates superior recommendations compared with state-of-the-art sequential recommendation approaches.

cs.AI↗

arXiv2021

Femtosecond dynamics of a polariton bosonic cascade at room temperature

Whispering gallery modes in a microwire are characterized by a nearly equidistant energy spectrum. In the strong exciton-photon coupling regime, this system represents a bosonic cascade: a ladder of discrete energy levels that sustains stimulated transitions between neighboring steps. In this work, by using femtosecond angle-resolved spectroscopic imaging technique, the ultrafast dynamics of polaritons in a bosonic cascade based on a one-dimensional ZnO whispering gallery microcavity is explicitly visualized. Clear ladder-form build-up process from higher to lower energy branches of the polariton condensates are observed, which are well reproduced by modeling using rate equations. Moreover, the polariton parametric scattering dynamics are distinguished on a timescale of hundreds of femtoseconds. Our understanding of the femtosecond condensation and scattering dynamics paves the way towards ultrafast coherent control of polaritons at room temperature, which will make it promising for high-speed all-optical integrated applications.

physics.optics↗

arXiv2021

The FAST Galactic Plane Pulsar Snapshot survey: I. Project design and pulsar discoveries

Discovery of pulsars is one of the main goals for large radio telescopes. The Five-hundred-meter Aperture Spherical radio Telescope (FAST), that incorporates an L-band 19-beam receiver with a system temperature of about 20~K, is the most sensitive radio telescope utilized for discovering pulsars. We designed the {\it snapshot} observation mode for a FAST key science project, the Galactic Plane Pulsar Snapshot (GPPS) survey, in which every four nearby pointings can observe {\it a cover} of a sky patch of 0.1575 square degrees through beam-switching of the L-band 19-beam receiver. The integration time for each pointing is 300 seconds so that the GPPS observations for a cover can be made in 21 minutes. The goal of the GPPS survey is to discover pulsars within the Galactic latitude of $\pm10^{\circ}$ from the Galactic plane, and the highest priority is given to the inner Galaxy within $\pm5^{\circ}$. Up to now, the GPPS survey has discovered 201 pulsars, including currently the faintest pulsars which cannot be detected by other telescopes, pulsars with extremely high dispersion measures (DMs) which challenge the currently widely used models for the Galactic electron density distribution, pulsars coincident with supernova remnants, 40 millisecond pulsars, 16 binary pulsars, some nulling and mode-changing pulsars and rotating radio transients (RRATs). The follow-up observations for confirmation of new pulsars have polarization-signals recorded for polarization profiles of the pulsars. Re-detection of previously known pulsars in the survey data also leads to significant improvements in parameters for 64 pulsars. The GPPS survey discoveries are published and will be updated at http://zmtt.bao.ac.cn/GPPS/ .

astro-ph.HE↗

arXiv2021

TCL: Transformer-based Dynamic Graph Modelling via Contrastive Learning

Dynamic graph modeling has recently attracted much attention due to its extensive applications in many real-world scenarios, such as recommendation systems, financial transactions, and social networks. Although many works have been proposed for dynamic graph modeling in recent years, effective and scalable models are yet to be developed. In this paper, we propose a novel graph neural network approach, called TCL, which deals with the dynamically-evolving graph in a continuous-time fashion and enables effective dynamic node representation learning that captures both the temporal and topology information. Technically, our model contains three novel aspects. First, we generalize the vanilla Transformer to temporal graph learning scenarios and design a graph-topology-aware transformer. Secondly, on top of the proposed graph transformer, we introduce a two-stream encoder that separately extracts representations from temporal neighborhoods associated with the two interaction nodes and then utilizes a co-attentional transformer to model inter-dependencies at a semantic level. Lastly, we are inspired by the recently developed contrastive learning and propose to optimize our model by maximizing mutual information (MI) between the predictive representations of two future interaction nodes. Benefiting from this, our dynamic representations can preserve high-level (or global) semantics about interactions and thus is robust to noisy interactions. To the best of our knowledge, this is the first attempt to apply contrastive learning to representation learning on dynamic graphs. We evaluate our model on four benchmark datasets for interaction prediction and experiment results demonstrate the superiority of our model.

cs.LG↗

arXiv2021

Towards an efficient framework for Data Extraction from Chart Images

In this paper, we fill the research gap by adopting state-of-the-art computer vision techniques for the data extraction stage in a data mining system. As shown in Fig.1, this stage contains two subtasks, namely, plot element detection and data conversion. For building a robust box detector, we comprehensively compare different deep learning-based methods and find a suitable method to detect box with high precision. For building a robust point detector, a fully convolutional network with feature fusion module is adopted, which can distinguish close points compared to traditional methods. The proposed system can effectively handle various chart data without making heuristic assumptions. For data conversion, we translate the detected element into data with semantic value. A network is proposed to measure feature similarities between legends and detected elements in the legend matching phase. Furthermore, we provide a baseline on the competition of Harvesting raw tables from Infographics. Some key factors have been found to improve the performance of each stage. Experimental results demonstrate the effectiveness of the proposed system.

cs.CV↗

arXiv2021

Well-posedness of the free boundary problem in incompressible MHD with surface tension

In this paper, we study the two phase flow problem with surface tension in the ideal incompressible magnetohydrodynamics. We first prove the local well-posedness of the two phase flow problem with surface tension, then demonstrate that as surface tension tends to zero, the solution of the two phase flow problem with surface tension converges to the solution of the two phase flow problem without surface tension.

math.AP↗

arXiv2021

Local spin alignment of vector mesons in relativistic heavy-ion collisions

We investigate the spin alignment of vector mesons arising from locally polarized quarks and anti-quarks (local spin alignment) in heavy-ion collisions. We find that $ρ_{00}\neq 1/3$ does not necessarily signal the global polarization of quarks and anti-quarks along the orbital angular momentum of the system, but may also originate from their local spin polarization. Such local spin polarization could be induced by local vorticity arising from the anisotropic expansion of the fireball. We further propose experimental observables that can distinguish between the local and the global spin alignments. Measurements of these observables in heavy-ion collisions can probe the vorticity pattern and shed light on the puzzles of $ϕ$ and $K^{*0}$ spin alignments.

nucl-th↗

arXiv2021

$\sin (ϕ_Λ-ϕ_S)$ azimuthal asymmetry in transversely polarized $Λ$ production in SIDIS within TMD factorization at EIC

We investigate the $\sin (ϕ_Λ-ϕ_S)$ single-spin asymmetry in the transversely polarized $Λ$ production in semi-inclusive deeply inelastic scattering process within the framework of the transverse momentum dependent~(TMD) factorization. The asymmetry is contributed by the convolution of the polarizing TMD fragmentation function $D_{1T}^\perp$ of the $Λ$ hyperon and the unpolarized TMD distribution function $f_1$ of the target nucleon. We apply two sets of $D_{1T}^{\perp,Λ^\uparrow/q}$, the spectator diquark model result and available parametrization, to numerically estimate the $\sin (ϕ_Λ-ϕ_S)$ asymmetry at the kinematical region of electron ion collider (EIC). We take into account the TMD evolution effect in order to include the scale dependence of the TMD distribution functions and fragmentation functions. It is found that different sets of $D_{1T}^{\perp,Λ^\uparrow/q}$ lead to different ${\sin (ϕ_Λ-ϕ_S)}$ asymmetry, particularly in sign. Future measurements on the $\sin (ϕ_Λ-ϕ_S)$ asymmetry with high precision at EIC can provide important cross check on the available $Λ$ polarizing fragmentation functions as well as constrain them more stringently.

hep-ph↗

arXiv2021

The transverse polarization of $Λ$ hyperons in $e^+e^-\toΛ^\uparrow h X$ processes within TMD factorization

We investigate the transverse polarization of the $Λ$ hyperon in the processes $e^+e^-\toΛ^\uparrow π^\pm X$ and $e^+e^-\toΛ^\uparrow K^\pm X$ within the framework of the transverse momentum dependent~(TMD) factorization. The transverse polarization is contributed by the convolution of the transversely polarizing fragmentation function~(PFF) $D_{1T}^\perp$ of the lambda hyperon and the unpolarized fragmentation function $D_1$ of pion/kaon. We adopt the spectator diquark model result for $D_{1T}^{\perp}$ to numerically estimate the transverse polarization in $e^+e^-\toΛ^\uparrow h X$ process at the kinematical region of Belle Collaboration. To implement the TMD evolution formalism of the fragmentation functions, we apply two different parametrizations on the nonperturbative Sudakov form factors associated with the fragmentation functions of the $Λ$, pion and kaon. It is found that our prediction on the polarization in the $Λπ^+$ production and $\barΛ π^-$ is consistent with the recent Belle measurement in size and sign, while the model predictions on the polarizations in $Λπ^-$ and $ΛK^\pm$ productions show strong disagreement with the Belle data. The reason for the discrepancies is discussed and possible approaches to improve the calculation in the future are also discussed.

hep-ph↗

arXiv2021

A Holistic Representation Guided Attention Network for Scene Text Recognition

Reading irregular scene text of arbitrary shape in natural images is still a challenging problem, despite the progress made recently. Many existing approaches incorporate sophisticated network structures to handle various shapes, use extra annotations for stronger supervision, or employ hard-to-train recurrent neural networks for sequence modeling. In this work, we propose a simple yet strong approach for scene text recognition. With no need to convert input images to sequence representations, we directly connect two-dimensional CNN features to an attention-based sequence decoder which guided by holistic representation. The holistic representation can guide the attention-based decoder focus on more accurate area. As no recurrent module is adopted, our model can be trained in parallel. It achieves 1.5x to 9.4x acceleration to backward pass and 1.3x to 7.9x acceleration to forward pass, compared with the RNN counterparts. The proposed model is trained with only word-level annotations. With this simple design, our method achieves state-of-the-art or competitive recognition performance on the evaluated regular and irregular scene text benchmark datasets.

cs.CV↗

arXiv2021

How does Working from Home Affect Developer Productivity? -- A Case Study of Baidu During COVID-19 Pandemic

Nowadays, working from home (WFH) has become a popular work arrangement due to its many potential benefits for both companies and employees (e.g., increasing job satisfaction and retention of employees). Many previous studies have investigated the impact of working from home on the productivity of employees. However, most of these studies usually use a qualitative analysis method such as survey and interview, and the studied participants do not work from home for a long continuing time. Due to the outbreak of coronavirus disease 2019 (COVID-19), a large number of companies asked their employees to work from home, which provides us an opportunity to investigate whether working from home affects their productivity. In this study, to investigate the difference of developer productivity between working from home and working onsite, we conduct a quantitative analysis based on a dataset of developers' daily activities from Baidu Inc, one of the largest IT companies in China. In total, we collected approximately four thousand records of 139 developers' activities of 138 working days. Out of these records, 1,103 records are submitted when developers work from home due to COVID-19 pandemic. We find that WFH has both positive and negative impacts on developer productivity in terms of different metrics, e.g., the number of builds/commits/code reviews. We also notice that working from home has different impacts on projects with different characteristics including programming language, project type/age/size. For example, working from home has a negative impact on developer productivity for large projects. Additionally, we find that productivity varies for different developers. Based on these findings, we get some feedbacks from developers of Baidu and understand some reasons why WFH has different impacts on developer productivity.

cs.SE↗

arXiv2021

Modulated MPC for Arm Inductor-less MVDC MMC with Reduced Computational Burden

A modulated model predictive controller is designed for an inductor-less modular multilevel converter targeting an MVDC solid-state transformer application. The underlying optimization problem is formulated such that a unique closed-form solution is derived from a highly accurate dynamic system model, which significantly reduces computation time. Unlike other closed-form methods, the proposed controller achieves inductor-less current control, has a reduced sampling speed enabled by a novel model-based circulating energy compensator, and does not require tuned PI controllers to generate current references. Simulation and experimental results demonstrate that the proposed controller achieves transient power flow control and steady state circulating energy control despite the low system inertia.

math.OC↗