Searcharxiv⌕ Search

arXiv subjects

Lei Wang

Publications and source records attributed to Lei Wang.

At least 721 records · Page 40Linked to original sources
arXiv2023

MoCoSA: Momentum Contrast for Knowledge Graph Completion with Structure-Augmented Pre-trained Language Models

Knowledge Graph Completion (KGC) aims to conduct reasoning on the facts within knowledge graphs and automatically infer missing links. Existing methods can mainly be categorized into structure-based or description-based. On the one hand, structure-based methods effectively represent relational facts in knowledge graphs using entity embeddings. However, they struggle with semantically rich real-world entities due to limited structural information and fail to generalize to unseen entities. On the other hand, description-based methods leverage pre-trained language models (PLMs) to understand textual information. They exhibit strong robustness towards unseen entities. However, they have difficulty with larger negative sampling and often lag behind structure-based methods. To address these issues, in this paper, we propose Momentum Contrast for knowledge graph completion with Structure-Augmented pre-trained language models (MoCoSA), which allows the PLM to perceive the structural information by the adaptable structure encoder. To improve learning efficiency, we proposed momentum hard negative and intra-relation negative sampling. Experimental results demonstrate that our approach achieves state-of-the-art performance in terms of mean reciprocal rank (MRR), with improvements of 2.5% on WN18RR and 21% on OpenBG500.

cs.CL↗
arXiv2023

New extended method for $ψ^{\prime}$ scaling function of inclusive electron scattering

Scaling analyses have been successfully applied to study the inclusive electron scattering $(\textit{e}, \textit{e}^{\prime})$ over the past few decades. In this paper, we utilize the $ψ^{\prime}$ scaling function in momentum space to analyze the $(\textit{e}, \textit{e}^{\prime})$ cross sections, where the nucleon momentum distributions are derived from self-consistent mean-field calculations. By further introducing the energy and momentum conservation in the scaling analysis, an improved $ψ^{\prime}$ scaling function is proposed to investigate the high-momentum part of the momentum distributions. Using the proposed scaling function, we systematically explore the effects of the nucleon-nucleon short-range correlation ($\textit{NN}$-SRC) on the $(\textit{e}, \textit{e}^{\prime})$ cross sections. From the experimental $(\textit{e}, \textit{e}^{\prime})$ data, the \textit{NN}-SRC strength is further extracted within the framework of the improved $ψ^{\prime}$ scaling function. The studies in this paper offer a new method to investigate the nucleon momentum distributions and the $\textit{NN}$-SRC effects in nuclei.

nucl-th↗
arXiv2023

Does AI for science need another ImageNet Or totally different benchmarks? A case study of machine learning force fields

AI for science (AI4S) is an emerging research field that aims to enhance the accuracy and speed of scientific computing tasks using machine learning methods. Traditional AI benchmarking methods struggle to adapt to the unique challenges posed by AI4S because they assume data in training, testing, and future real-world queries are independent and identically distributed, while AI4S workloads anticipate out-of-distribution problem instances. This paper investigates the need for a novel approach to effectively benchmark AI for science, using the machine learning force field (MLFF) as a case study. MLFF is a method to accelerate molecular dynamics (MD) simulation with low computational cost and high accuracy. We identify various missed opportunities in scientifically meaningful benchmarking and propose solutions to evaluate MLFF models, specifically in the aspects of sample efficiency, time domain sensitivity, and cross-dataset generalization capabilities. By setting up the problem instantiation similar to the actual scientific applications, more meaningful performance metrics from the benchmark can be achieved. This suite of metrics has demonstrated a better ability to assess a model's performance in real-world scientific applications, in contrast to traditional AI benchmarking methodologies. This work is a component of the SAIBench project, an AI4S benchmarking suite. The project homepage is https://www.computercouncil.org/SAIBench.

cs.LG↗
arXiv2023

Automatic Radiology Report Generation by Learning with Increasingly Hard Negatives

Automatic radiology report generation is challenging as medical images or reports are usually similar to each other due to the common content of anatomy. This makes a model hard to capture the uniqueness of individual images and is prone to producing undesired generic or mismatched reports. This situation calls for learning more discriminative features that could capture even fine-grained mismatches between images and reports. To achieve this, this paper proposes a novel framework to learn discriminative image and report features by distinguishing them from their closest peers, i.e., hard negatives. Especially, to attain more discriminative features, we gradually raise the difficulty of such a learning task by creating increasingly hard negative reports for each image in the feature space during training, respectively. By treating the increasingly hard negatives as auxiliary variables, we formulate this process as a min-max alternating optimisation problem. At each iteration, conditioned on a given set of hard negative reports, image and report features are learned as usual by minimising the loss functions related to report generation. After that, a new set of harder negative reports will be created by maximising a loss reflecting image-report alignment. By solving this optimisation, we attain a model that can generate more specific and accurate reports. It is noteworthy that our framework enhances discriminative feature learning without introducing extra network weights. Also, in contrast to the existing way of generating hard negatives, our framework extends beyond the granularity of the dataset by generating harder samples out of the training set. Experimental study on benchmark datasets verifies the efficacy of our framework and shows that it can serve as a plug-in to readily improve existing medical report generation models.

cs.CV↗
arXiv2023

Differentiable programming tensor networks for Kitaev magnets

We present a general computational framework to investigate ground state properties of quantum spin models on infinite two-dimensional lattices using automatic differentiation-based gradient optimization of infinite projected entangled-pair states. The approach exploits the variational uniform matrix product states to contract infinite tensor networks with unit-cell structure and incorporates automatic differentiation to optimize the local tensors. We applied this framework to the Kitaev-type model, which involves complex interactions and competing ground states. To evaluate the accuracy of this method, we compared the results with exact solutions for the Kitaev model and found that it has a better agreement for various observables compared to previous tensor network calculations based on imaginary-time projection. Additionally, by finding out the ground state with lower variational energy compared to previous studies, we provided convincing evidence for the existence of nematic paramagnetic phases and 18-site configuration in the phase diagram of the $K$-$Γ$ model. Furthermore, in the case of the realistic $K$-$J$-$Γ$-$Γ'$ model for the Kitaev material $α$-RuCl$_3$, we discovered a non-colinear zigzag ground state. Lastly, we also find that the strength of the critical out-of-plane magnetic field that suppresses such a zigzag state has a lower transition field value than the previous finite-cylinder calculations. The framework is versatile and will be useful for a quick scan of phase diagrams for a broad class of quantum spin models.

physics.comp-ph↗
arXiv2023

Hierarchical Spatio-Temporal Representation Learning for Gait Recognition

Gait recognition is a biometric technique that identifies individuals by their unique walking styles, which is suitable for unconstrained environments and has a wide range of applications. While current methods focus on exploiting body part-based representations, they often neglect the hierarchical dependencies between local motion patterns. In this paper, we propose a hierarchical spatio-temporal representation learning (HSTL) framework for extracting gait features from coarse to fine. Our framework starts with a hierarchical clustering analysis to recover multi-level body structures from the whole body to local details. Next, an adaptive region-based motion extractor (ARME) is designed to learn region-independent motion features. The proposed HSTL then stacks multiple ARMEs in a top-down manner, with each ARME corresponding to a specific partition level of the hierarchy. An adaptive spatio-temporal pooling (ASTP) module is used to capture gait features at different levels of detail to perform hierarchical feature mapping. Finally, a frame-level temporal aggregation (FTA) module is employed to reduce redundant information in gait sequences through multi-scale temporal downsampling. Extensive experiments on CASIA-B, OUMVLP, GREW, and Gait3D datasets demonstrate that our method outperforms the state-of-the-art while maintaining a reasonable balance between model accuracy and complexity.

cs.CV↗
arXiv2023

IterLara: A Turing Complete Algebra for Big Data, AI, Scientific Computing, and Database

\textsc{Lara} is a key-value algebra that aims at unifying linear and relational algebra with three types of operation abstraction. The study of \textsc{Lara}'s expressive ability reports that it can represent relational algebra and most linear algebra operations. However, several essential computations, such as matrix inversion and determinant, cannot be expressed in \textsc{Lara}. \textsc{Lara} cannot represent global and iterative computation, either. This article proposes \textsc{IterLara}, extending \textsc{Lara} with iterative operators, to provide an algebraic model that unifies operations in general-purpose computing, like big data, AI, scientific computing, and database. We study the expressive ability of \textsc{Lara} and \textsc{IterLara} and prove that \textsc{IterLara} with aggregation functions can represent matrix inversion, determinant. Besides, we demonstrate that \textsc{IterLara} with no limitation of function utility is Turing complete. We also propose the Operation Count (OP) as a metric of computation amount for \textsc{IterLara} and ensure that the OP metric is in accordance with the existing computation metrics.

cs.DB↗
arXiv2023

DBFed: Debiasing Federated Learning Framework based on Domain-Independent

As digital transformation continues, enterprises are generating, managing, and storing vast amounts of data, while artificial intelligence technology is rapidly advancing. However, it brings challenges in information security and data security. Data security refers to the protection of digital information from unauthorized access, damage, theft, etc. throughout its entire life cycle. With the promulgation and implementation of data security laws and the emphasis on data security and data privacy by organizations and users, Privacy-preserving technology represented by federated learning has a wide range of application scenarios. Federated learning is a distributed machine learning computing framework that allows multiple subjects to train joint models without sharing data to protect data privacy and solve the problem of data islands. However, the data among multiple subjects are independent of each other, and the data differences in quality may cause fairness issues in federated learning modeling, such as data bias among multiple subjects, resulting in biased and discriminatory models. Therefore, we propose DBFed, a debiasing federated learning framework based on domain-independent, which mitigates model bias by explicitly encoding sensitive attributes during client-side training. This paper conducts experiments on three real datasets and uses five evaluation metrics of accuracy and fairness to quantify the effect of the model. Most metrics of DBFed exceed those of the other three comparative methods, fully demonstrating the debiasing effect of DBFed.

cs.LG↗
arXiv2023

PIMCOMP: A Universal Compilation Framework for Crossbar-based PIM DNN Accelerators

Crossbar-based PIM DNN accelerators can provide massively parallel in-situ operations. A specifically designed compiler is important to achieve high performance for a wide variety of DNN workloads. However, some key compilation issues such as parallelism considerations, weight replication selection, and array mapping methods have not been solved. In this work, we propose PIMCOMP - a universal compilation framework for NVM crossbar-based PIM DNN accelerators. PIMCOMP is built on an abstract PIM accelerator architecture, which is compatible with the widely used Crossbar/IMA/Tile/Chip hierarchy. On this basis, we propose four general compilation stages for crossbar-based PIM accelerators: node partitioning, weight replicating, core mapping, and dataflow scheduling. We design two compilation modes with different inter-layer pipeline granularities to support high-throughput and low-latency application scenarios, respectively. Our experimental results show that PIMCMOP yields improvements of 1.6$\times$ and 2.4$\times$ in throughput and latency, respectively, relative to PUMA.

cs.AR↗
arXiv2023

An Empirical Study on the Holiday Effect of China's Time-Honored Companies

The stock segment of China's time-honored brand enterprises has an important position in our securities stock market. The holiday effect is one of the market anomalies that occur in the securities market, which refers to the phenomenon that the stock market has significantly different returns than other trading days around festivals. The study of the holiday effect of China's time-honored brand enterprises can provide fresh ideas for the revitalization of our time-honored brands and the revitalization of time-honored enterprises. This paper takes listed companies of China's time-honored brand enterprises as the research object and focuses on the impact of the holiday effect on listed companies of China's time-honored brands with the help of the event study, and empirically analyses the changes in the return of listed companies of China time-honored brands during the Spring Festival period from 2012 to 2021. The empirical results reveal that: the time-honored brand concept stocks have a significant post-holiday effect during the Chinese New Year period, the time-honored alcoholic beverage enterprises are more sensitive to the Chinese New Year reflection, while the holiday effect of the time-honored pharmaceutical manufacturing enterprises is not significant.

q-fin.GN↗
arXiv2023

Determination of geopotential difference by hydrogen masers based on precise point positioning time-frequency transfer

According to the general relativity theory, the geopotential difference can be determined by gravity frequency shift between two clocks. Here we report on the experiments to determine the geopotential difference between two remote sites by hydrogen masers based on precise point positioning time-frequency transfer technique. The experiments include the remote clock comparison and the local clock comparison using two CH1-95 active hydrogen masers linked with global navigation satellite system time-frequency receivers. The frequency difference between two hydrogen masers at two sites is derived from the time difference series resolved by the above-mentioned technique. Considering the local clock comparison as calibration, the determined geopotential difference by our experiments is 12,142.3 (112.4) m^2/s^2, quite close to the value 12,153.3 (2.3) m^2/s^2 computed by the EIGEN-6C4 model. Results show that the proposed approach here for determining geopotential difference is feasible, operable, and promising for applications in various fields.

physics.geo-ph↗
arXiv2023

GaitMM: Multi-Granularity Motion Sequence Learning for Gait Recognition

Gait recognition aims to identify individual-specific walking patterns by observing the different periodic movements of each body part. However, most existing methods treat each part equally and fail to account for the data redundancy caused by the different step frequencies and sampling rates of gait sequences. In this study, we propose a multi-granularity motion representation network (GaitMM) for gait sequence learning. In GaitMM, we design a combined full-body and fine-grained sequence learning module (FFSL) to explore part-independent spatio-temporal representations. Moreover, we utilize a frame-wise compression strategy, referred to as multi-scale motion aggregation (MSMA), to capture discriminative information in the gait sequence. Experiments on two public datasets, CASIA-B and OUMVLP, show that our approach reaches state-of-the-art performances.

cs.CV↗
arXiv2023

PEBO-SLAM: Observer design for visual inertial SLAM with convergence guarantees

This paper introduces a new linear parameterization to the problem of visual inertial simultaneous localization and mapping (VI-SLAM) -- without any approximation -- for the case only using information from a single monocular camera and an inertial measurement unit. In this problem set, the system state evolves on the nonlinear manifold $SE(3)\times \mathbb{R}^{3n}$, on which we design dynamic extensions carefully to generate invariant foliations, such that the problem can be reformulated into online \emph{constant parameter} identification, then interestingly with linear regression models obtained. It demonstrates that VI-SLAM can be translated into a linear least squares problem, in the deterministic sense, \emph{globally} and \emph{exactly}. Based on this observation, we propose a novel SLAM observer, following the recently established parameter estimation-based observer (PEBO) methodology. A notable merit is that the proposed observer enjoys almost global asymptotic stability, requiring neither persistency of excitation nor uniform complete observability, which, however, are widely adopted in most existing works with provable stability but can hardly be assured in many practical scenarios.

eess.SY↗
arXiv2023

$m^\ast$ of two-dimensional electron gas: a neural canonical transformation study

The quasiparticle effective mass $m^\ast$ of interacting electrons is a fundamental quantity in the Fermi liquid theory. However, the precise value of the effective mass of uniform electron gas is still elusive after decades of research. The newly developed neural canonical transformation approach [Xie et al., J. Mach. Learn. 1, (2022)] offers a principled way to extract the effective mass of electron gas by directly calculating the thermal entropy at low temperature. The approach models a variational many-electron density matrix using two generative neural networks: an autoregressive model for momentum occupation and a normalizing flow for electron coordinates. Our calculation reveals a suppression of effective mass in the two-dimensional spin-polarized electron gas, which is more pronounced than previous reports in the low-density strong-coupling region. This prediction calls for verification in two-dimensional electron gas experiments.

cond-mat.stat-mech↗
arXiv2023

Legion: Automatically Pushing the Envelope of Multi-GPU System for Billion-Scale GNN Training

Graph neural network(GNN) has been widely applied in real-world applications, such as product recommendation in e-commerce platforms and risk control in financial management systems. Several cache-based GNN systems have been built to accelerate GNN training in a single machine with multiple GPUs. However, these systems fail to train billion-scale graphs efficiently, which is a common challenge in the industry. In this work, we propose Legion, a system that automatically pushes the envelope of multi-GPU systems for accelerating billion-scale GNN training. First, we design a hierarchical graph partitioning mechanism that significantly improves the multi-GPU cache performance. Second, we build a unified multi-GPU cache that helps to minimize the PCIe traffic incurred by caching both graph topology and features with the highest hotness. Third, we develop an automatic caching management mechanism that adapts the multi-GPU cache plan according to the hardware specifications and various graphs to maximize the overall training throughput. Evaluations on various GNN models and multiple datasets show that Legion supports training billion-scale GNNs in a single machine and significantly outperforms the state-of-the-art cache-based systems on small graphs.

cs.DC↗
arXiv2023

AD-NEGF: An End-to-End Differentiable Quantum Transport Simulator for Sensitivity Analysis and Inverse Problems

Since proposed in the 70s, the Non-Equilibrium Green Function (NEGF) method has been recognized as a standard approach to quantum transport simulations. Although it achieves superiority in simulation accuracy, the tremendous computational cost makes it unbearable for high-throughput simulation tasks such as sensitivity analysis, inverse design, etc. In this work, we propose AD-NEGF, to our best knowledge the first end-to-end differentiable NEGF model for quantum transport simulations. We implement the entire numerical process in PyTorch, and design customized backward pass with implicit layer techniques, which provides gradient information at an affordable cost while guaranteeing the correctness of the forward simulation. The proposed model is validated with applications in calculating differential physical quantities, empirical parameter fitting, and doping optimization, which demonstrates its capacity to accelerate the material design process by conducting gradient-based parameter optimization.

cond-mat.mes-hall↗
arXiv2023

Do-GOOD: Towards Distribution Shift Evaluation for Pre-Trained Visual Document Understanding Models

Numerous pre-training techniques for visual document understanding (VDU) have recently shown substantial improvements in performance across a wide range of document tasks. However, these pre-trained VDU models cannot guarantee continued success when the distribution of test data differs from the distribution of training data. In this paper, to investigate how robust existing pre-trained VDU models are to various distribution shifts, we first develop an out-of-distribution (OOD) benchmark termed Do-GOOD for the fine-Grained analysis on Document image-related tasks specifically. The Do-GOOD benchmark defines the underlying mechanisms that result in different distribution shifts and contains 9 OOD datasets covering 3 VDU related tasks, e.g., document information extraction, classification and question answering. We then evaluate the robustness and perform a fine-grained analysis of 5 latest VDU pre-trained models and 2 typical OOD generalization algorithms on these OOD datasets. Results from the experiments demonstrate that there is a significant performance gap between the in-distribution (ID) and OOD settings for document images, and that fine-grained analysis of distribution shifts can reveal the brittle nature of existing pre-trained VDU models and OOD generalization algorithms. The code and datasets for our Do-GOOD benchmark can be found at https://github.com/MAEHCM/Do-GOOD.

cs.CV↗
arXiv2023

Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models

Large language models (LLMs) have recently been shown to deliver impressive performance in various NLP tasks. To tackle multi-step reasoning tasks, few-shot chain-of-thought (CoT) prompting includes a few manually crafted step-by-step reasoning demonstrations which enable LLMs to explicitly generate reasoning steps and improve their reasoning task accuracy. To eliminate the manual effort, Zero-shot-CoT concatenates the target problem statement with "Let's think step by step" as an input prompt to LLMs. Despite the success of Zero-shot-CoT, it still suffers from three pitfalls: calculation errors, missing-step errors, and semantic misunderstanding errors. To address the missing-step errors, we propose Plan-and-Solve (PS) Prompting. It consists of two components: first, devising a plan to divide the entire task into smaller subtasks, and then carrying out the subtasks according to the plan. To address the calculation errors and improve the quality of generated reasoning steps, we extend PS prompting with more detailed instructions and derive PS+ prompting. We evaluate our proposed prompting strategy on ten datasets across three reasoning problems. The experimental results over GPT-3 show that our proposed zero-shot prompting consistently outperforms Zero-shot-CoT across all datasets by a large margin, is comparable to or exceeds Zero-shot-Program-of-Thought Prompting, and has comparable performance with 8-shot CoT prompting on the math reasoning problem. The code can be found at https://github.com/AGI-Edgerunners/Plan-and-Solve-Prompting.

cs.CL↗