arXiv subjects

Lei Chen

Publications and source records attributed to Lei Chen.

At least 487 records · Page 27Linked to original sources

arXiv2022

GridTuner: Reinvestigate Grid Size Selection for Spatiotemporal Prediction Models [Technical Report]

With the development of traffic prediction technology, spatiotemporal prediction models have attracted more and more attention from academia communities and industry. However, most existing researches focus on reducing model's prediction error but ignore the error caused by the uneven distribution of spatial events within a region. In this paper, we study a region partitioning problem, namely optimal grid size selection problem (OGSS), which aims to minimize the real error of spatiotemporal prediction models by selecting the optimal grid size. In order to solve OGSS, we analyze the upper bound of real error of spatiotemporal prediction models and minimize the real error by minimizing its upper bound. Through in-depth analysis, we find that the upper bound of real error will decrease then increase when the number of model grids increase from 1 to the maximum allowed value. Then, we propose two algorithms, namely Ternary Search and Iterative Method, to automatically find the optimal grid size. Finally, the experiments verify that the error of prediction has the same trend as its upper bound, and the change trend of the upper bound of real error with respect to the increase of the number of model grids will decrease then increase. Meanwhile, in a case study, by selecting the optimal grid size, the order dispatching results of a state-of-the-art prediction-based algorithm can be improved up to 13.6%, which shows the effectiveness of our methods on tuning the region partition for spatiotemporal prediction models.

cs.DB↗

arXiv2021

Multiorbital spin-triplet pairing and spin resonance in the heavy-fermion superconductor $\mathrm{UTe_2}$

The heavy-fermion system $\mathrm{UTe_2}$ is a candidate for spin-triplet superconductivity, which is of considerable interest to quantum engineering. Among the outstanding issues is the nature of the pairing state. A recent surprising discovery is the observation of a resonance in the spin excitation spectrum at an antiferromagnetic wavevector [C. Duan {\it et al.}, Nature \textbf{600}, 636 (2021)], which stands in apparent contrast to the ferromagnetic nature of the interactions expected in this system. We show how the puzzle can be resolved by a multiorbital spin-triplet pairing constructed from local degrees of freedom. Because it does not commute with the kinetic part of the Hamiltonian, the pairing contains both intra- and inter-band terms in the band basis. We demonstrate that the intraband pairing component naturally yields a spin resonance at the antiferromagnetic wavevector. Our work illustrates how orbital degrees of freedom can enrich the nature and properties of spin-triplet superconductivity of strongly-correlated quantum materials.

cond-mat.supr-con↗

arXiv2021

Constraining mapping class group homomorphisms using finite subgroups

We classify homomorphisms from mapping class groups by using finite subgroups. First, we give a new proof of a result of Aramayona--Souto that homomorphisms between mapping class groups of closed surfaces are trivial for a range of genera. Second, we show that only finitely many mapping class groups of closed surfaces have non-trivial homomorphisms into $\text{Homeo}(\mathbb{S}^n)$ for any $n$. We also prove that every homomorphism from $\text{Mod}(S_g)$ to $\text{Homeo}(\mathbb{S}^2)$ or $\text{Homeo}(\mathbb{S}^3)$ is trivial if $g\ge 3$, extending a result of Franks--Handel.

math.GT↗

arXiv2021

An Adaptive Graph Pre-training Framework for Localized Collaborative Filtering

Graph neural networks (GNNs) have been widely applied in the recommendation tasks and have obtained very appealing performance. However, most GNN-based recommendation methods suffer from the problem of data sparsity in practice. Meanwhile, pre-training techniques have achieved great success in mitigating data sparsity in various domains such as natural language processing (NLP) and computer vision (CV). Thus, graph pre-training has the great potential to alleviate data sparsity in GNN-based recommendations. However, pre-training GNNs for recommendations face unique challenges. For example, user-item interaction graphs in different recommendation tasks have distinct sets of users and items, and they often present different properties. Therefore, the successful mechanisms commonly used in NLP and CV to transfer knowledge from pre-training tasks to downstream tasks such as sharing learned embeddings or feature extractors are not directly applicable to existing GNN-based recommendations models. To tackle these challenges, we delicately design an adaptive graph pre-training framework for localized collaborative filtering (ADAPT). It does not require transferring user/item embeddings, and is able to capture both the common knowledge across different graphs and the uniqueness for each graph. Extensive experimental results have demonstrated the effectiveness and superiority of ADAPT.

cs.IR↗

arXiv2021

Semi-Supervised Variational User Identity Linkage via Noise-Aware Self-Learning

User identity linkage, which aims to link identities of a natural person across different social platforms, has attracted increasing research interest recently. Existing approaches usually first embed the identities as deterministic vectors in a shared latent space, and then learn a classifier based on the available annotations. However, the formation and characteristics of real-world social platforms are full of uncertainties, which makes these deterministic embedding based methods sub-optimal. In addition, it is intractable to collect sufficient linkage annotations due to the tremendous gaps between different platforms. Semi-supervised models utilize the unlabeled data to help capture the intrinsic data distribution, which are more promising in practical usage. However, the existing semi-supervised linkage methods heavily rely on the heuristically defined similarity measurements to incorporate the innate closeness between labeled and unlabeled samples. Such manually designed assumptions may not be consistent with the actual linkage signals and further introduce the noises. To address the mentioned limitations, in this paper we propose a novel Noise-aware Semi-supervised Variational User Identity Linkage (NSVUIL) model. Specifically, we first propose a novel supervised linkage module to incorporate the available annotations. Each social identity is represented by a Gaussian distribution in the Wasserstein space to simultaneously preserve the fine-grained social profiles and model the uncertainty of identities. Then, a noise-aware self-learning module is designed to faithfully augment the few available annotations, which is capable of filtering noises from the pseudo-labels generated by the supervised module.

cs.SI↗

arXiv2021

AutoGEL: An Automated Graph Neural Network with Explicit Link Information

Recently, Graph Neural Networks (GNNs) have gained popularity in a variety of real-world scenarios. Despite the great success, the architecture design of GNNs heavily relies on manual labor. Thus, automated graph neural network (AutoGNN) has attracted interest and attention from the research community, which makes significant performance improvements in recent years. However, existing AutoGNN works mainly adopt an implicit way to model and leverage the link information in the graphs, which is not well regularized to the link prediction task on graphs, and limits the performance of AutoGNN for other graph tasks. In this paper, we present a novel AutoGNN work that explicitly models the link information, abbreviated to AutoGEL. In such a way, AutoGEL can handle the link prediction task and improve the performance of AutoGNNs on the node classification and graph classification task. Specifically, AutoGEL proposes a novel search space containing various design dimensions at both intra-layer and inter-layer designs and adopts a more robust differentiable search algorithm to further improve efficiency and effectiveness. Experimental results on benchmark data sets demonstrate the superiority of AutoGEL on several tasks.

cs.LG↗

arXiv2021

Persia: An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters

Deep learning based models have dominated the current landscape of production recommender systems. Furthermore, recent years have witnessed an exponential growth of the model scale--from Google's 2016 model with 1 billion parameters to the latest Facebook's model with 12 trillion parameters. Significant quality boost has come with each jump of the model capacity, which makes us believe the era of 100 trillion parameters is around the corner. However, the training of such models is challenging even within industrial scale data centers. This difficulty is inherited from the staggering heterogeneity of the training computation--the model's embedding layer could include more than 99.99% of the total model size, which is extremely memory-intensive; while the rest neural network is increasingly computation-intensive. To support the training of such huge models, an efficient distributed training system is in urgent need. In this paper, we resolve this challenge by careful co-design of both the optimization algorithm and the distributed system architecture. Specifically, in order to ensure both the training efficiency and the training accuracy, we design a novel hybrid training algorithm, where the embedding layer and the dense neural network are handled by different synchronization mechanisms; then we build a system called Persia (short for parallel recommendation training system with hybrid acceleration) to support this hybrid training algorithm. Both theoretical demonstration and empirical study up to 100 trillion parameters have conducted to justified the system design and implementation of Persia. We make Persia publicly available (at https://github.com/PersiaML/Persia) so that anyone would be able to easily train a recommender model at the scale of 100 trillion parameters.

cs.LG↗

arXiv2021

Chaotic dynamics of Bose-Einstein condensate in a density-dependent gauge field

In this work we study the effect of density-dependent gauge field on the collective dynamics of a harmonically trapped Bose-Einstein condensate, beyond the linear response regime. The densitydependent gauge field, as a backaction of the condensate, can in turn affect the condensate dynamics, resulting in highly nonlinear equations of motion. We find that the dipole and breathing oscillations of the condensate along the direction of gauge field are coupled by this field. For a quasi-onedimensional condensate, this coupling makes the collective motion quasiperiodic. While for a quasitwo-dimensional condensate, the gauge field can also induce a Hall effect, manifested as an additional coupling between dipole and breathing oscillations in perpendicular direction. When the densitydependent gauge field is strong, the interplay between these oscillations can cause the collective dynamics of the condensate to become chaotic. Our findings reveal an important effect of dynamical gauge field on the nonlinear dynamics of a Bose-Einstein condensate.

cond-mat.quant-gas↗

arXiv2021

Electron correlations and $T$-breaking density wave order in a $\mathbb{Z}_2$ kagome metal

There have been extensive recent developments on kagome metals, such as T$_m$X$_n$ (T= Fe, Co and X= Sn, Ge) and $A$V$_3$Sb$_5$ ($A=$ Cs, K, Rb). An emerging issue is the nature of correlated phases when topologically \textit{non-trivial} bands cross the Fermi level. Here, we consider an extended Hubbard model on the kagome lattice in the presence of spin-orbit couplings, involving a Kramer's pair of bands that have opposite Chern numbers and are isolated in the band structure. We construct an effective model in a time-reversal (T) symmetric lattice description. We determine the correlated phases of this model and identify a density-wave order in the phase diagram. We show that this order is T-breaking, which originates from the Wannier orbitals lacking a common Wannier center -- a fingerprint of the underlying $Z_2$ topology. Implications of our results for the correlation physics of the kagome metals are discussed.

cond-mat.str-el↗

arXiv2021

Solving Partial Differential Equations with Point Source Based on Physics-Informed Neural Networks

In recent years, deep learning technology has been used to solve partial differential equations (PDEs), among which the physics-informed neural networks (PINNs) emerges to be a promising method for solving both forward and inverse PDE problems. PDEs with a point source that is expressed as a Dirac delta function in the governing equations are mathematical models of many physical processes. However, they cannot be solved directly by conventional PINNs method due to the singularity brought by the Dirac delta function. We propose a universal solution to tackle this problem with three novel techniques. Firstly the Dirac delta function is modeled as a continuous probability density function to eliminate the singularity; secondly a lower bound constrained uncertainty weighting algorithm is proposed to balance the PINNs losses between point source area and other areas; and thirdly a multi-scale deep neural network with periodic activation function is used to improve the accuracy and convergence speed of the PINNs method. We evaluate the proposed method with three representative PDEs, and the experimental results show that our method outperforms existing deep learning-based methods with respect to the accuracy, the efficiency and the versatility.

cs.LG↗

arXiv2021

MixSeq: Connecting Macroscopic Time Series Forecasting with Microscopic Time Series Data

Time series forecasting is widely used in business intelligence, e.g., forecast stock market price, sales, and help the analysis of data trend. Most time series of interest are macroscopic time series that are aggregated from microscopic data. However, instead of directly modeling the macroscopic time series, rare literature studied the forecasting of macroscopic time series by leveraging data on the microscopic level. In this paper, we assume that the microscopic time series follow some unknown mixture probabilistic distributions. We theoretically show that as we identify the ground truth latent mixture components, the estimation of time series from each component could be improved because of lower variance, thus benefitting the estimation of macroscopic time series as well. Inspired by the power of Seq2seq and its variants on the modeling of time series data, we propose Mixture of Seq2seq (MixSeq), an end2end mixture model to cluster microscopic time series, where all the components come from a family of Seq2seq models parameterized by different parameters. Extensive experiments on both synthetic and real-world data show the superiority of our approach.

cs.LG↗

arXiv2021

A SVD-Based Synchrophasor Estimator for P-class PMUs with Improved Immune from Interharmonic Tones

The increasing use of renewable generation, power electronic devices, and nonlinear loads in power systems brings more severe interharmonic tones to the measurand, which can increase estimation errors of P-class PMUs, cause misoperation of protection relays, and even threaten the stability of the power systems. Therefore, the performance of the P-class synchrophasor estimator under interharmonic interference should be evaluated and new estimation schemes that can improve the operational robustness are needed for various protection and control applications. In this paper, a synchrophasor estimator design for P-class PMUs that introduces singular value decomposition to the least square algorithm based on the Taylor series is proposed. By constructing an optimization with proposed adjustable parameters, finite impulse response filters with high transition band attenuation are designed. Numerical and experimental tests verify the proposed dynamic synchrophasor estimator performance and show an effective improvement in rejecting interharmonic tones, especially when a short time window and light computational burden are required.

physics.ins-det↗

arXiv2021

Towards More Effective and Economic Sparsely-Activated Model

The sparsely-activated models have achieved great success in natural language processing through large-scale parameters and relatively low computational cost, and gradually become a feasible technique for training and implementing extremely large models. Due to the limit of communication cost, activating multiple experts is hardly affordable during training and inference. Therefore, previous work usually activate just one expert at a time to alleviate additional communication cost. Such routing mechanism limits the upper bound of model performance. In this paper, we first investigate a phenomenon that increasing the number of activated experts can boost the model performance with higher sparse ratio. To increase the number of activated experts without an increase in computational cost, we propose SAM (Switch and Mixture) routing, an efficient hierarchical routing mechanism that activates multiple experts in a same device (GPU). Our methods shed light on the training of extremely large sparse models and experiments prove that our models can achieve significant performance gain with great efficiency improvement.

cs.CL↗

arXiv2021

Memory-Efficient CNN Accelerator Based on Interlayer Feature Map Compression

Existing deep convolutional neural networks (CNNs) generate massive interlayer feature data during network inference. To maintain real-time processing in embedded systems, large on-chip memory is required to buffer the interlayer feature maps. In this paper, we propose an efficient hardware accelerator with an interlayer feature compression technique to significantly reduce the required on-chip memory size and off-chip memory access bandwidth. The accelerator compresses interlayer feature maps through transforming the stored data into frequency domain using hardware-implemented 8x8 discrete cosine transform (DCT). The high-frequency components are removed after the DCT through quantization. Sparse matrix compression is utilized to further compress the interlayer feature maps. The on-chip memory allocation scheme is designed to support dynamic configuration of the feature map buffer size and scratch pad size according to different network-layer requirements. The hardware accelerator combines compression, decompression, and CNN acceleration into one computing stream, achieving minimal compressing and processing delay. A prototype accelerator is implemented on an FPGA platform and also synthesized in TSMC 28-nm COMS technology. It achieves 403GOPS peak throughput and 1.4x~3.3x interlayer feature map reduction by adding light hardware area overhead, making it a promising hardware accelerator for intelligent IoT devices.

cs.AR↗

arXiv2021

Effect of Visual Cues on Pointing Tasks in Co-located Augmented Reality Collaboration

Visual cues are essential in computer-mediated communication. It is especially important when communication happens in a collaboration scenario that requires focusing several users' attention on aspecific object among other similar ones. This paper explores the effect of visual cues on pointing tasks in co-located Augmented Reality (AR) collaboration. A user study (N = 32, 16 pairs) was conducted to compare two types of visual cues: Pointing Line (PL)and Moving Track (MT). Both are head-based visual techniques.Through a series of collaborative pointing tasks on objects with different states (static and dynamic) and density levels (low, mediumand high), the results showed that PL was better on task performance and usability, but MT was rated higher on social presenceand user preference. Based on our results, some design implicationsare provided for pointing tasks in co-located AR collaboration.

cs.HC↗

arXiv2021

Spin-singlet and spin-triplet pairing correlations in antiferromagnetically coupled Kondo systems

Recent experiments in quantum critical heavy fermion metals have pushed to the fore the question about whether antiferromagnetic fluctuations can promote both spin-singlet and spin-triplet superconductivity. Here we address the issue through non-perturbative calculations in antiferromagnetically correlated Kondo systems. We identify Kondo-destruction quantum critical points in both the SU(2) symmetric and Ising-anisotropic cluster Bose-Fermi Anderson models. The spin-singlet pairing correlations are significantly enhanced near the quantum critical point in the SU(2) case; however, with adequate but still realistic degree of Ising anisotropy, the spin-triplet pairing correlations are competitive. Our results demonstrate that spin-flip processes strengthen the spin-singlet pairing at the Kondo-destruction quantum critical points, and point towards a way for antiferromagnetic correlations to drive spin-triplet pairing. Further implications of our findings in the broader context of strongly correlated superconductivity are discussed.

cond-mat.str-el↗

arXiv2021

Unconventional Superconductivity from Fermi Surface Fluctuations in Strongly Correlated Metals

In quantum materials, electrons that have strong correlations tend to localize, leading to quantum spins as the building blocks for low-energy physics. When strongly correlated electrons coexist with more weakly-correlated conduction electrons, multiple channels of effective interactions develop and compete with each other. The competition creates quantum fluctuations having a large spectral weight, with the associated entropies reaching significant fractions of $R\ln 2$ per electron. Advancing a framework to understand how the fluctuating local moments influence unconventional superconductivity is both pressing and challenging. Here we do so in the exemplary setting of heavy-fermion metals, where the amplified quantum fluctuations manifest in the form of Kondo destruction and large-to-small Fermi-surface fluctuations. These fluctuations lead to unconventional superconductivity whose transition temperature is exceptionally high relative to the effective Fermi temperature, reaching several percent of the Kondo temperature scale. Our results provide a natural understanding of the enigmatic superconductivity in a host of heavy-fermion metals. Moreover, the qualitative physics underlying our findings and their implications for the formation of unconventional superconductivity apply to a variety of highly correlated metals with strong Fermi surface fluctuations.

cond-mat.str-el↗

arXiv2021

Fine-Grained Element Identification in Complaint Text of Internet Fraud

Existing system dealing with online complaint provides a final decision without explanations. We propose to analyse the complaint text of internet fraud in a fine-grained manner. Considering the complaint text includes multiple clauses with various functions, we propose to identify the role of each clause and classify them into different types of fraud element. We construct a large labeled dataset originated from a real finance service platform. We build an element identification model on top of BERT and propose additional two modules to utilize the context of complaint text for better element label classification, namely, global context encoder and label refiner. Experimental results show the effectiveness of our model.

cs.CL↗