arXiv subjects

Pengfei Wang

Publications and source records attributed to Pengfei Wang.

At least 163 records · Page 9Linked to original sources

arXiv2023

MultiEM: Efficient and Effective Unsupervised Multi-Table Entity Matching

Entity Matching (EM), which aims to identify all entity pairs referring to the same real-world entity from relational tables, is one of the most important tasks in real-world data management systems. Due to the labeling process of EM being extremely labor-intensive, unsupervised EM is more applicable than supervised EM in practical scenarios. Traditional unsupervised EM assumes that all entities come from two tables; however, it is more common to match entities from multiple tables in practical applications, that is, multi-table entity matching (multi-table EM). Unfortunately, effective and efficient unsupervised multi-table EM remains under-explored. To fill this gap, this paper formally studies the problem of unsupervised multi-table entity matching and proposes an effective and efficient solution, termed as MultiEM. MultiEM is a parallelable pipeline of enhanced entity representation, table-wise hierarchical merging, and density-based pruning. Extensive experimental results on six real-world benchmark datasets demonstrate the superiority of MultiEM in terms of effectiveness and efficiency.

cs.DB↗

arXiv2023

Style Interleaved Learning for Generalizable Person Re-identification

Domain generalization (DG) for person re-identification (ReID) is a challenging problem, as access to target domain data is not permitted during the training process. Most existing DG ReID methods update the feature extractor and classifier parameters based on the same features. This common practice causes the model to overfit to existing feature styles in the source domain, resulting in sub-optimal generalization ability on target domains. To solve this problem, we propose a novel style interleaved learning (IL) framework. Unlike conventional learning strategies, IL incorporates two forward propagations and one backward propagation for each iteration. We employ the features of interleaved styles to update the feature extractor and classifiers using different forward propagations, which helps to prevent the model from overfitting to certain domain styles. To generate interleaved feature styles, we further propose a new feature stylization approach. It produces a wide range of meaningful styles that are both different and independent from the original styles in the source domain, which caters to the IL methodology. Extensive experimental results show that our model not only consistently outperforms state-of-the-art methods on large-scale benchmarks for DG ReID, but also has clear advantages in computational efficiency. The code is available at https://github.com/WentaoTan/Interleaved-Learning.

cs.CV↗

arXiv2023

Soliton generation in CaF$_2$ crystalline whispering gallery mode resonators with negative thermal-optical effects

Calcium fluoride (CaF$_2$) crystalline whispering gallery mode resonators (WGMRs) exhibit ultrahigh intrinsic quality factors and a low power anomalous dispersion in the communication and mid-infrared bands, making them attractive platforms for microresonator-based comb generation. However, their unique negative thermo-optic effects pose challenges when achieving thermal equilibrium. To our knowledge, our experiments serve as the first demonstration of soliton microcombs in Q > 109 CaF$_2$ WGMRs. We observed soliton mode-locking and bidirectional switching of soliton numbers caused by the negative thermo-optic effects. Additionally, various soliton formation dynamics are shown, including breathing and vibrational solitons, which can be attributed to thermo-photomechanical oscillations. Thus, our results enrich the soliton generation platform and provide a reference for generating solitons from WGMRs that comprise other materials with negative thermo-optic effects. In the future, the ultrahigh quality factor of CaF$_2$ crystal cavities may enable the generation of sub-milliwatt-level broad-spectrum soliton combs.

physics.optics↗

arXiv2023

Sensing orbital hybridization of graphene-diamond interface with a single spin

Interfacial interactions are crucial in a variety of fields and can greatly affect the electric, magnetic, and chemical properties of materials. Among them, interface orbital hybridization plays a fundamental role in the properties of surface electrons such as dispersion, interaction, and ground states. Conventional measurements of electronic states at interfaces such as scanning tunneling microscopes are all based on electric interactions which, however, suffer from strong perturbation on these electrons. Here we unveil a new experimental detection of interface electrons based on the weak magnetic interactions between them and the nitrogen-vacancy (NV) center in diamond. With negligible perturbation on the interface electrons, their physical properties can be revealed by the NV spin coherence time. In our system, the interface interaction leads to significant decreases in both the density and coherence time of the electron spins at the diamond-graphene interface. Furthermore, together with electron spin resonance spectra and first-principle calculations, we can retrieve the effect of interface electron orbital hybridization. Our study opens a new pathway toward the microscopic probing of interfacial electronic states with weak magnetic interactions and provides a new avenue for future research on material interfaces.

quant-ph↗

arXiv2023

Evaporation characteristics of Er$^{3+}$ doped silica fiber and its application in the preparation of whispering gallery mode lasers

The fabrication of whispering gallery lasers (WGL) is used to experimentally evaluate the evaporation rate (mol/$μ$m) and ratio (mol/mol) of erbium and silica lost from a doped fiber during heating. Fixed lengths of doped silica fiber are spliced to different lengths of undoped fiber and then evaporated by feeding into the focus of a CO$_{2}$ laser. During evaporation, erbium ions are precipitated in the doped silica fiber to control the erbium concentration in the remaining SiO$_2$, which is melted into a microsphere. By increasing the length of the undoped section, a critical point is reached where effectively no ions remain in the glass microsphere. The critical point is found using the lasing spectra of the whispering gallery modes in microspheres with equal sizes. From the critical point, it is estimated that, for a given CO$_{2}$ laser power, $6.36 \times 10^{-21}$~mol of Er$^{3+}$ is lost during the evaporation process for every cubic micron of silica fiber. This is equivalent to $1.74 \times 10^{-7}$~mol of Er$^{3+}$ lost per mol of SiO$_{2}$ evaporated. This result facilitates the control of the doping concentration in WGLs and provides insight into the kinetics of laser-induced evaporation of doped silica.

physics.optics↗

arXiv2023

Convergence of the number of period sets in strings

Consider words of length $n$. The set of all periods of a word of length $n$ is a subset of $\{0,1,2,\ldots,n-1\}$. However, any subset of $\{0,1,2,\ldots,n-1\}$ is not necessarily a valid set of periods. In a seminal paper in 1981, Guibas and Odlyzko have proposed to encode the set of periods of a word into an $n$ long binary string, called an autocorrelation, where a one at position $i$ denotes the period $i$. They considered the question of recognizing a valid period set, and also studied the number of valid period sets for length $n$, denoted $κ_n$. They conjectured that $\ln(κ_n)$ asymptotically converges to a constant times $\ln^2(n)$. If improved lower bounds for $\ln(κ_n)/\ln^2(n)$ were proposed in 2001, the question of a tight upper bound has remained opened since Guibas and Odlyzko's paper. Here, we exhibit an upper bound for this fraction, which implies its convergence and closes this long standing conjecture. Moreover, we extend our result to find similar bounds for the number of correlations: a generalization of autocorrelations which encodes the overlaps between two strings.

cs.DM↗

arXiv2023

Adaptive Path-Memory Network for Temporal Knowledge Graph Reasoning

Temporal knowledge graph (TKG) reasoning aims to predict the future missing facts based on historical information and has gained increasing research interest recently. Lots of works have been made to model the historical structural and temporal characteristics for the reasoning task. Most existing works model the graph structure mainly depending on entity representation. However, the magnitude of TKG entities in real-world scenarios is considerable, and an increasing number of new entities will arise as time goes on. Therefore, we propose a novel architecture modeling with relation feature of TKG, namely aDAptivE path-MemOry Network (DaeMon), which adaptively models the temporal path information between query subject and each object candidate across history time. It models the historical information without depending on entity representation. Specifically, DaeMon uses path memory to record the temporal path information derived from path aggregation unit across timeline considering the memory passing strategy between adjacent timestamps. Extensive experiments conducted on four real-world TKG datasets demonstrate that our proposed model obtains substantial performance improvement and outperforms the state-of-the-art up to 4.8% absolute in MRR.

cs.AI↗

arXiv2023

Modeling Entities as Semantic Points for Visual Information Extraction in the Wild

Recently, Visual Information Extraction (VIE) has been becoming increasingly important in both the academia and industry, due to the wide range of real-world applications. Previously, numerous works have been proposed to tackle this problem. However, the benchmarks used to assess these methods are relatively plain, i.e., scenarios with real-world complexity are not fully represented in these benchmarks. As the first contribution of this work, we curate and release a new dataset for VIE, in which the document images are much more challenging in that they are taken from real applications, and difficulties such as blur, partial occlusion, and printing shift are quite common. All these factors may lead to failures in information extraction. Therefore, as the second contribution, we explore an alternative approach to precisely and robustly extract key information from document images under such tough conditions. Specifically, in contrast to previous methods, which usually either incorporate visual information into a multi-modal architecture or train text spotting and information extraction in an end-to-end fashion, we explicitly model entities as semantic points, i.e., center points of entities are enriched with semantic information describing the attributes and relationships of different entities, which could largely benefit entity labeling and linking. Extensive experiments on standard benchmarks in this field as well as the proposed dataset demonstrate that the proposed method can achieve significantly enhanced performance on entity labeling and linking, compared with previous state-of-the-art models. Dataset is available at https://www.modelscope.cn/datasets/damo/SIBR/summary.

cs.CV↗

arXiv2023

Sharpness-Aware Gradient Matching for Domain Generalization

The goal of domain generalization (DG) is to enhance the generalization capability of the model learned from a source domain to other unseen domains. The recently developed Sharpness-Aware Minimization (SAM) method aims to achieve this goal by minimizing the sharpness measure of the loss landscape. Though SAM and its variants have demonstrated impressive DG performance, they may not always converge to the desired flat region with a small loss value. In this paper, we present two conditions to ensure that the model could converge to a flat minimum with a small loss, and present an algorithm, named Sharpness-Aware Gradient Matching (SAGM), to meet the two conditions for improving model generalization capability. Specifically, the optimization objective of SAGM will simultaneously minimize the empirical risk, the perturbed loss (i.e., the maximum loss within a neighborhood in the parameter space), and the gap between them. By implicitly aligning the gradient directions between the empirical risk and the perturbed loss, SAGM improves the generalization capability over SAM and its variants without increasing the computational cost. Extensive experimental results show that our proposed SAGM method consistently outperforms the state-of-the-art methods on five DG benchmarks, including PACS, VLCS, OfficeHome, TerraIncognita, and DomainNet. Codes are available at https://github.com/Wang-pengfei/SAGM.

cs.CV↗

arXiv2023

Reaction plane alignment with linearly polarized photon in heavy-ion collisions

The collective observables play critical roles in probing the properties of quark-gluon-plasma created in relativistic heavy-ion collisions, in which the information on initial collision geometry is crucial. However, the initial collision geometry, e.g., the reaction plane, cannot be directly extracted in the experiment. In this paper, we demonstrate the idea of determining the reaction plane via the feature of linear polarization of the coherent photoproduction process and discuss the advantages of the proposed approach in comparison with traditional methods.

hep-ph↗

arXiv2023

Candidate Soups: Fusing Candidate Results Improves Translation Quality for Non-Autoregressive Translation

Non-autoregressive translation (NAT) model achieves a much faster inference speed than the autoregressive translation (AT) model because it can simultaneously predict all tokens during inference. However, its translation quality suffers from degradation compared to AT. And existing NAT methods only focus on improving the NAT model's performance but do not fully utilize it. In this paper, we propose a simple but effective method called "Candidate Soups," which can obtain high-quality translations while maintaining the inference speed of NAT models. Unlike previous approaches that pick the individual result and discard the remainders, Candidate Soups (CDS) can fully use the valuable information in the different candidate translations through model uncertainty. Extensive experiments on two benchmarks (WMT'14 EN-DE and WMT'16 EN-RO) demonstrate the effectiveness and generality of our proposed method, which can significantly improve the translation quality of various base models. More notably, our best variant outperforms the AT model on three translation tasks with 7.6 times speedup.

cs.CL↗

arXiv2022

Traceable Automatic Feature Transformation via Cascading Actor-Critic Agents

Feature transformation for AI is an essential task to boost the effectiveness and interpretability of machine learning (ML). Feature transformation aims to transform original data to identify an optimal feature space that enhances the performances of a downstream ML model. Existing studies either combines preprocessing, feature selection, and generation skills to empirically transform data, or automate feature transformation by machine intelligence, such as reinforcement learning. However, existing studies suffer from: 1) high-dimensional non-discriminative feature space; 2) inability to represent complex situational states; 3) inefficiency in integrating local and global feature information. To fill the research gap, we formulate the feature transformation task as an iterative, nested process of feature generation and selection, where feature generation is to generate and add new features based on original features, and feature selection is to remove redundant features to control the size of feature space. Finally, we present extensive experiments and case studies to illustrate 24.7\% improvements in F1 scores compared with SOTAs and robustness in high-dimensional data.

cs.LG↗

arXiv2022

Two-sample Behrens--Fisher problems for high-dimensional data: a normal reference F-type test

The problem of testing the equality of mean vectors for high-dimensional data has been intensively investigated in the literature. However, most of the existing tests impose strong assumptions on the underlying group covariance matrices which may not be satisfied or hardly be checked in practice. In this article, an F-type test for two-sample Behrens--Fisher problems for high-dimensional data is proposed and studied. When the two samples are normally distributed and when the null hypothesis is valid, the proposed F-type test statistic is shown to be an F-type mixture, a ratio of two independent chi-square-type mixtures. Under some regularity conditions and the null hypothesis, it is shown that the proposed F-type test statistic and the above F-type mixture have the same normal and non-normal limits. It is then justified to approximate the null distribution of the proposed F-type test statistic by that of the F-type mixture, resulting in the so-called normal reference F-type test. Since the F-type mixture is a ratio of two independent chi-square-type mixtures, we employ the Welch--Satterthwaite chi-square-approximation to the distributions of the numerator and the denominator of the F-type mixture respectively, resulting in an approximation F-distribution whose degrees of freedom can be consistently estimated from the data. The asymptotic power of the proposed F-type test is established. Two simulation studies are conducted and they show that in terms of size control, the proposed F-type test outperforms two existing competitors. The proposed F-type test is also illustrated by a real data example.

math.ST↗

arXiv2022

Boosting Urban Traffic Speed Prediction via Integrating Implicit Spatial Correlations

Urban traffic speed prediction aims to estimate the future traffic speed for improving the urban transportation services. Enormous efforts have been made on exploiting spatial correlations and temporal dependencies of traffic speed evolving patterns by leveraging explicit spatial relations (geographical proximity) through pre-defined geographical structures ({\it e.g.}, region grids or road networks). While achieving promising results, current traffic speed prediction methods still suffer from ignoring implicit spatial correlations (interactions), which cannot be captured by grid/graph convolutions. To tackle the challenge, we propose a generic model for enabling the current traffic speed prediction methods to preserve implicit spatial correlations. Specifically, we first develop a Dual-Transformer architecture, including a Spatial Transformer and a Temporal Transformer. The Spatial Transformer automatically learns the implicit spatial correlations across the road segments beyond the boundary of geographical structures, while the Temporal Transformer aims to capture the dynamic changing patterns of the implicit spatial correlations. Then, to further integrate both explicit and implicit spatial correlations, we propose a distillation-style learning framework, in which the existing traffic speed prediction methods are considered as the teacher model, and the proposed Dual-Transformer architectures are considered as the student model. The extensive experiments over three real-world datasets indicate significant improvements of our proposed framework over the existing methods.

cs.LG↗

arXiv2022

Multiple Testing in Genome-Wide Association Studies via Hierarchical Hidden Markov Models

The problems of large-scale multiple testing are often encountered in modern scientific researches. Conventional multiple testing procedures usually suffer considerable loss of testing efficiency due to the lack of consideration of correlations among tests. In fact, the appropriate use of correlation information not only enhances the efficacy of multiple testing but also improves the interpretability of the results. Since the disease- or trait-related single nucleotide polymorphisms (SNPs) often tend to be clustered and exhibit serial correlations, the hidden Markov model (HMM) based multiple testing procedure has been successfully applied in genome-wide association studies (GWAS). It is important to note that modeling the entire chromosome using one HMM is somewhat rough. To overcome this issue, this paper employs the hierarchical hidden Markov model (HHMM) to describe local correlations among tests and develops a multiple testing procedure that can not only automatically divide different class of chromosome regions, but also takes into account local correlations among tests. Theoretically, it is shown that the proposed multiple testing procedure is valid and optimal in some sense. Then a data-driven procedure is developed to mimic the oracle version. Extensive simulations and the real data analysis show that the novel multiple testing procedure outperforms its competitors.

stat.ME↗

arXiv2022

SurfaceVoronoi: Efficiently Computing Voronoi Diagrams over Mesh Surfaces with Arbitrary Distance Solvers

In this paper, we propose to compute Voronoi diagrams over mesh surfaces driven by an arbitrary geodesic distance solver, assuming that the input is a triangle mesh as well as a collection of sites $P=\{p_i\}_{i=1}^m$ on the surface. We propose two key techniques to solve this problem. First, as the partition is determined by minimizing the $m$ distance fields, each of which rooted at a source site, we suggest keeping one or more distance triples, for each triangle, that may help determine the Voronoi bisectors when one uses a mark-and-sweep geodesic algorithm to predict the multi-source distance field. Second, rather than keep the distance itself at a mesh vertex, we use the squared distance to characterize the linear change of distance field restricted in a triangle, which is proved to induce an exact VD when the base surface reduces to a planar triangle mesh. Specially, our algorithm also supports the Euclidean distance, which can handle thin-sheet models (e.g. leaf) and runs faster than the traditional restricted Voronoi diagram~(RVD) algorithm. It is very extensible to deal with various variants of surface-based Voronoi diagrams including (1)surface-based power diagram, (2)constrained Voronoi diagram with curve-type breaklines, and (3)curve-type generators. We conduct extensive experimental results to validate the ability to approximate the exact VD in different distance-driven scenarios.

cs.CG↗

arXiv2022

Peering into the Milky Way by FAST: III. Magnetic fields in the Galactic halo and farther spiral arms revealed by the Faraday effect of faint pulsars

The Five-hundred-meter Aperture Spherical radio Telescope (FAST) is the most sensitive radio telescope for pulsar observations. We make polarimetric measurements of a large number of faint and distant pulsars using the FAST. We present the new measurements of Faraday rotation for 134 faint pulsars in the Galactic halo. Significant improvements are also made for some basic pulsar parameters for 15 of them. We analyse the newly determined rotation measures (RMs) for the Galactic magnetic fields by using these 134 halo pulsars, together with previously available RMs for pulsars and extragalactic radio sources and also the newly determined RMs for another 311 faint pulsars which are either newly discovered in the project of the Galactic Plane Pulsar Snapshot (GPPS) survey or previously known pulsars without RMs. The RM tomographic analysis in the first Galactic quadrant gives roughly the same field strength of around 2~$μ$G for the large-scale toroidal halo magnetic fields. The scale height of the halo magnetic fields is found to be at least 2.7$\pm$0.3~kpc. The RM differentiation of a large number of pulsars in the Galactic disk in the Galactic longitude range of $26^{\circ}<l<90^{\circ}$ gives evidence for the clockwise magnetic fields (viewed from the north Galactic pole) in two interarm regions inside the Scutum arm and between the Scutum and Sagittarius arm, and the clockwise fields in the Local-Perseus interarm region and field reversals in the Perseus arm and beyond.

astro-ph.GA↗

arXiv2022

Cascadable in-memory computing based on symmetric writing and read out

The building block of in-memory computing with spintronic devices is mainly based on the magnetic tunnel junction with perpendicular interfacial anisotropy (p-MTJ). The resulting asymmetric write and read-out operations impose challenges in downscaling and direct cascadability of p-MTJ devices. Here, we propose that a new symmetric write and read-out mechanism can be realized in perpendicular-anisotropy spin-orbit (PASO) quantum materials based on Fe3GeTe2 and WTe2. We demonstrate that field-free and deterministic reversal of the perpendicular magnetization can be achieved by employing unconventional charge to z-spin conversion. The resulting magnetic state can be readily probed with its intrinsic inverse process, i.e., z-spin to charge conversion. Using the PASO quantum material as a fundamental building block, we implement the functionally complete set of logic-in-memory operations and a more complex nonvolatile half-adder logic function. Our work highlights the potential of PASO quantum materials for the development of scalable energy-efficient and ultrafast spintronic computing.

cond-mat.mes-hall↗