arXiv subjects

Lei Chen

Publications and source records attributed to Lei Chen.

At least 433 records · Page 24Linked to original sources

arXiv2023

A Fast Insertion Operator for Ridesharing over Time-Dependent Road Networks

Ridesharing has become a promising travel mode recently due to the economic and social benefits. As an essential operator, "insertion operator" has been extensively studied over static road networks. When a new request appears, the insertion operator is used to find the optimal positions of a worker's current route to insert the origin and destination of this request and minimize the travel time of this worker. Previous works study how to conduct the insertion operation efficiently in static road networks, however, in reality, the route planning should be addressed by considering the dynamic traffic scenario (i.e., a time-dependent road network). Unfortunately, existing solutions to the insertion operator become in efficient under this setting. Thus, this paper studies the insertion operator over time-dependent road networks. Specially, to reduce the high time complexity $O(n^3)$ of existing solution, we calculate the compound travel time functions along the route to speed up the calculation of the travel time between vertex pairs belonging to the route, as a result time complexity of an insertion can be reduced to $O(n^2)$. Finally, we further improve the method to a linear-time insertion algorithm by showing that it only needs $O(1)$ time to find the best position of current route to insert the origin when linearly enumerating each possible position for the new request's destination. Evaluations on two real-world and large-scale datasets show that our methods can accelerate the existing insertion algorithm by up to 25 times.

cs.DB↗

arXiv2023

Querying Shortest Path on Large Time-Dependent Road Networks with Shortcuts

Querying the shortest path between two vertexes is a fundamental operation in a variety of applications, which has been extensively studied over static road networks. However, in reality, the travel costs of road segments evolve over time, and hence the road network can be modeled as a time-dependent graph. In this paper, we study the shortest path query over large-scale time-dependent road networks. Existing work focuses on a hierarchical partition structure, which makes the index construction and travel cost query inefficient. To improve the efficiency of such queries, we propose a novel index by decomposing a road network into a tree structure and selecting a set of shortcuts on the tree to speed up the query processing. Specifically, we first formally define a shortcut selection problem over the tree decomposition of the time-dependent road network. This problem, which is proven to be NP-hard, aims to select and build the most effective shortcut set. We first devise a dynamic programming method with exact results to solve the selection problem. To obtain the optimal shortcut set quickly, we design an approximation algorithm that guarantees a 0.5-approximation ratio. Based on the novel tree structure, we devise a shortcut-based algorithm to answer the shortest path query over time-dependent road networks. Finally, we conduct extensive performance studies using large-scale real-world road networks. The results demonstrate that our method can achieve better efficiency and scalability than the state-of-the-art method.

cs.DB↗

arXiv2023

Accurate and Efficient Trajectory-based Contact Tracing with Secure Computation and Geo-Indistinguishability

Contact tracing has been considered as an effective measure to limit the transmission of infectious disease such as COVID-19. Trajectory-based contact tracing compares the trajectories of users with the patients, and allows the tracing of both direct contacts and indirect contacts. Although trajectory data is widely considered as sensitive and personal data, there is limited research on how to securely compare trajectories of users and patients to conduct contact tracing with excellent accuracy, high efficiency, and strong privacy guarantee. Traditional Secure Multiparty Computation (MPC) techniques suffer from prohibitive running time, which prevents their adoption in large cities with millions of users. In this work, we propose a technical framework called ContactGuard to achieve accurate, efficient, and privacy-preserving trajectory-based contact tracing. It improves the efficiency of the MPC-based baseline by selecting only a small subset of locations of users to compare against the locations of the patients, with the assist of Geo-Indistinguishability, a differential privacy notion for Location-based services (LBS) systems. Extensive experiments demonstrate that ContactGuard runs up to 2.6$\times$ faster than the MPC baseline, with no sacrifice in terms of the accuracy of contact tracing.

cs.DB↗

arXiv2023

Efficient and Accurate Range Counting on Privacy-preserving Spatial Data Federation [Technical Report]

A spatial data federation is a collection of data owners (e.g., a consortium of taxi companies), and collectively it could provide better location-based services (LBS). For example, car-hailing services over a spatial data federation allow end users to easily pick the best offers. We focus on the range counting queries, which are primitive operations in spatial databases but received little attention in related research, especially considering the privacy requirements from data owners, who are reluctant to disclose their proprietary data. We propose a grouping-based technical framework named FedGroup, which groups data owners without compromising privacy, and achieves superior query accuracy (up to 50% improvement) as compared to directly applying existing privacy mechanisms achieving Differential Privacy (DP). Our experimental results also demonstrate that FedGroup runs orders-of-magnitude faster than traditional Secure Multiparty Computation (MPC) based method, and FedGroup even scales to millions of data owners, which is a common setting in the era of ubiquitous mobile devices.

cs.DB↗

arXiv2023

Time-aware Dynamic Graph Embedding for Asynchronous Structural Evolution

Dynamic graphs refer to graphs whose structure dynamically changes over time. Despite the benefits of learning vertex representations (i.e., embeddings) for dynamic graphs, existing works merely view a dynamic graph as a sequence of changes within the vertex connections, neglecting the crucial asynchronous nature of such dynamics where the evolution of each local structure starts at different times and lasts for various durations. To maintain asynchronous structural evolutions within the graph, we innovatively formulate dynamic graphs as temporal edge sequences associated with joining time of vertices (ToV) and timespan of edges (ToE). Then, a time-aware Transformer is proposed to embed vertices' dynamic connections and ToEs into the learned vertex representations. Meanwhile, we treat each edge sequence as a whole and embed its ToV of the first vertex to further encode the time-sensitive information. Extensive evaluations on several datasets show that our approach outperforms the state-of-the-art in a wide range of graph mining tasks. At the same time, it is very efficient and scalable for embedding large-scale dynamic graphs.

cs.LG↗

arXiv2023

FTSO: Effective NAS via First Topology Second Operator

Existing one-shot neural architecture search (NAS) methods have to conduct a search over a giant super-net, which leads to the huge computational cost. To reduce such cost, in this paper, we propose a method, called FTSO, to divide the whole architecture search into two sub-steps. Specifically, in the first step, we only search for the topology, and in the second step, we search for the operators. FTSO not only reduces NAS's search time from days to 0.68 seconds, but also significantly improves the found architecture's accuracy. Our extensive experiments on ImageNet show that within 18 seconds, FTSO can achieve a 76.4% testing accuracy, 1.5% higher than the SOTA, PC-DARTS. In addition, FTSO can reach a 97.77% testing accuracy, 0.27% higher than the SOTA, with nearly 100% (99.8%) search time saved, when searching on CIFAR10.

cs.CV↗

arXiv2023

Observation of Kondo lattice and Kondo-enhanced anomalous Hall effect in an itinerant ferromagnet

The interplay between Kondo screening and magnetic interactions is central to comprehending the intricate phases in heavy-fermion compounds. However, the role of the itinerant magnetic order, which is driven by the conducting (c) electrons, has been largely uncharted in the context of heavy-fermion systems due to the scarcity of material candidates. Here we demonstrate the coexistence of the coherent Kondo screening and d-orbital ferromagnetism in material system La$_{1-x}$Ce$_x$Co$_2$As$_2$, through comprehensive thermodynamic and electrical transport measurements. Additionally, using angle-resolved photoemission spectroscopy (ARPES), we further observe the f-orbit-dominated bands near the Fermi level ($E_f$) and signatures of the f-c hybridization below the magnetic transition temperature, providing strong evidence of Kondo lattice state in the presence of ferromagnetic order. Remarkably, by changing the ratio of Ce/La, we observe a substantial enhancement of the anomalous Hall effect (AHE) in the Kondo lattice regime. The value of the Hall conductivity quantitatively matches with the first-principle calculation that optimized with our ARPES results and can be attributed to the large Berry curvature (BC) density engendered by the topological nodal rings composed of the Ce-4f and Co-3d orbitals at $E_f$. Our findings point to the realization of a new platform for exploring correlation-driven topological responses in a novel Kondo lattice environment.

cond-mat.str-el↗

arXiv2023

Improving Recommendation Fairness via Data Augmentation

Collaborative filtering based recommendation learns users' preferences from all users' historical behavior data, and has been popular to facilitate decision making. R Recently, the fairness issue of recommendation has become more and more essential. A recommender system is considered unfair when it does not perform equally well for different user groups according to users' sensitive attributes~(e.g., gender, race). Plenty of methods have been proposed to alleviate unfairness by optimizing a predefined fairness goal or changing the distribution of unbalanced training data. However, they either suffered from the specific fairness optimization metrics or relied on redesigning the current recommendation architecture. In this paper, we study how to improve recommendation fairness from the data augmentation perspective. The recommendation model amplifies the inherent unfairness of imbalanced training data. We augment imbalanced training data towards balanced data distribution to improve fairness. The proposed framework is generally applicable to any embedding-based recommendation, and does not need to pre-define a fairness metric. Extensive experiments on two real-world datasets clearly demonstrate the superiority of our proposed framework. We publish the source code at https://github.com/newlei/FDA.

cs.IR↗

arXiv2023

On the equivalence between graph isomorphism testing and function approximation with GNNs

Graph Neural Networks (GNNs) have achieved much success on graph-structured data. In light of this, there have been increasing interests in studying their expressive power. One line of work studies the capability of GNNs to approximate permutation-invariant functions on graphs, and another focuses on the their power as tests for graph isomorphism. Our work connects these two perspectives and proves their equivalence. We further develop a framework of the expressive power of GNNs that incorporates both of these viewpoints using the language of sigma-algebra, through which we compare the expressive power of different types of GNNs together with other graph isomorphism tests. In particular, we prove that the second-order Invariant Graph Network fails to distinguish non-isomorphic regular graphs with the same degree. Then, we extend it to a new architecture, Ring-GNN, which succeeds in distinguishing these graphs and achieves good performances on real-world datasets.

cs.LG↗

arXiv2023

JWST/MIRI Spectroscopy of the Disk of the Young Eruptive Star EX Lup in Quiescence

EX Lup is a low-mass pre-main sequence star that occasionally shows accretion-related outbursts. Here, we present JWST/MIRI medium resolution spectroscopy obtained for EX Lup fourteen years after its powerful outburst. EX Lup is now in quiescence and displays a Class II spectrum. We detect a forest of emission lines from molecules previously identified in infrared spectra of classical T Tauri disks: H2O, OH, H2, HCN, C2H2, and CO2. The detection of organic molecules demonstrates that they are back after disappearing during the large outburst. Spectral lines from water and OH are for the first time de-blended and will provide a much improved characterization of their distribution and density in the inner disk. The spectrum also shows broad emission bands from warm, sub-micron size amorphous silicate grains at 10 and 18 um. During the outburst, in 2008, crystalline forsterite grains were annealed in the inner disk within 1 au, but their spectral signatures in the 10 um silicate band later disappeared. With JWST we re-discovered these crystals via their 19.0, 20.0, and 23.5 um emission, whose strength implies that the particles are at ~3 au from the star. This suggests that crystalline grains formed in 2008 were transported outwards and now approach the water snowline, where they may be incorporated into planetesimals. Containing several key tracers of planetesimal and planet formation, EX Lup is an ideal laboratory to study the effects of variable luminosity on the planet-forming material and may provide explanation for the observed high crystalline fraction in solar system comets.

astro-ph.SR↗

arXiv2023

Metastable supersolid in spin-orbit coupled Bose-Einstein condensates

Supersolid is a special state of matter with both superfluid properties and spontaneous modulation of particle density. In this paper, we focus on the supersolid stripe phase realized in a spin-orbit coupled Bose-Einstein condensate and explore the properties of a class of metastable supersolids. In particular, we study a one-dimensional supersolid whose characteristic wave number $k$ (magnitude of wave vector) deviates from $k_{m}$, i.e., the one at ground state. In other words, the period of density modulation is shorter or longer than the one at ground state. We find that this class of supersolids can still be stable if their wave numbers fall in the range $k_{c1}<k<k_{c2}$, with two thresholds $k_{c1}$ and $k_{c2}$. Stripes with $k$ outside this range suffer from dynamical instability with complex Bogoliubov excitation spectrum at long wavelength. Experimentally, these stripes with $k$ away from $k_m$ are accessible by exciting the longitudinal spin dipole mode, resulting in temporal oscillation of stripe period as well as $k$. Within the mean-field Gross-Pitaevskii theory, we numerically confirm that for a large enough amplitude of spin dipole oscillation, the stripe states become unstable through breaking periodicity, in qualitative agreement with the existence of thresholds of $k$ for stable stripes. Our work extends the concept of supersolid and uncovers a new class of metastable supersolids to explore.

cond-mat.quant-gas↗

arXiv2023

Mapping class groups of circle bundles over a surface

In this paper, we study the algebraic structure of mapping class group $Mod(X)$ of 3-manifolds $X$ that fiber as a circle bundle over a surface $S^1\rightarrow X\rightarrow S_g$. There is an exact sequence $1\rightarrow H^1(S_g)\rightarrow Mod(X)\rightarrow Mod(S_g)\rightarrow1$. We relate this to the Birman exact sequence and determine when this sequence splits.

math.GT↗

arXiv2022

Meta-Auto-Decoder for Solving Parametric Partial Differential Equations

Many important problems in science and engineering require solving the so-called parametric partial differential equations (PDEs), i.e., PDEs with different physical parameters, boundary conditions, shapes of computation domains, etc. Recently, building learning-based numerical solvers for parametric PDEs has become an emerging new field. One category of methods such as the Deep Galerkin Method (DGM) and Physics-Informed Neural Networks (PINNs) aim to approximate the solution of the PDEs. They are typically unsupervised and mesh-free, but require going through the time-consuming network training process from scratch for each set of parameters of the PDE. Another category of methods such as Fourier Neural Operator (FNO) and Deep Operator Network (DeepONet) try to approximate the solution mapping directly. Being fast with only one forward inference for each PDE parameter without retraining, they often require a large corpus of paired input-output observations drawn from numerical simulations, and most of them need a predefined mesh as well. In this paper, we propose Meta-Auto-Decoder (MAD), a mesh-free and unsupervised deep learning method that enables the pre-trained model to be quickly adapted to equation instances by implicitly encoding (possibly heterogenous) PDE parameters as latent vectors. The proposed method MAD can be interpreted by manifold learning in infinite-dimensional spaces, granting it a geometric insight. Extensive numerical experiments show that the MAD method exhibits faster convergence speed without losing accuracy than other deep learning-based methods. The project page with code is available: https://gitee.com/mindspore/mindscience/tree/master/MindElec/.

cs.LG↗

arXiv2022

The Highest Melting Point Material: Searched by Bayesian Global Optimization with Deep Potential Molecular Dynamics

The interest in refractory materials is increasing rapidly in recent decades due to the development of hypersonic vehicles. However, which substance has the highest melting point keeps a secret, since precise measurements in extreme condition are overwhelmingly difficult. In the present work, an accurate deep potential model of Hf-Ta-C-N system was firstly trained, and then applied to search for the highest melting point material by using molecular dynamics simulation and Bayesian global optimization. The predicted melting points agree well with experiments, and confirm that the carbon site vacancy can enhance melting points of rock-salt structure carbides. Solid solution with N is verified as another new and more effective melting point enhancing approach for HfC, while the conventional routing of solid solution with Ta (e.g. HfTa4C5) is not suggested to result in a maximum melting point. The highest melting point (~ 4236 K) is achieved with composition of HfC0.638N0.271, which is ~ 80 K higher than the highest value in Hf-C binary system. The dominating mechanism of N addition is believed to be the instable C-N and N-N bonds in liquid phase, which reduces the liquid phase entropy and renders the liquid phase less stable. The improved melting point and fewer gas generation during oxidation by addition of N provides new routing to modify the thermal protection materials for hypersonic vehicles.

cond-mat.mtrl-sci↗

arXiv2022

MSRL: Distributed Reinforcement Learning with Dataflow Fragments

Reinforcement learning (RL) trains many agents, which is resource-intensive and must scale to large GPU clusters. Different RL training algorithms offer different opportunities for distributing and parallelising the computation. Yet, current distributed RL systems tie the definition of RL algorithms to their distributed execution: they hard-code particular distribution strategies and only accelerate specific parts of the computation (e.g. policy network updates) on GPU workers. Fundamentally, current systems lack abstractions that decouple RL algorithms from their execution. We describe MindSpore Reinforcement Learning (MSRL), a distributed RL training system that supports distribution policies that govern how RL training computation is parallelised and distributed on cluster resources, without requiring changes to the algorithm implementation. MSRL introduces the new abstraction of a fragmented dataflow graph, which maps Python functions from an RL algorithm's training loop to parallel computational fragments. Fragments are executed on different devices by translating them to low-level dataflow representations, e.g. computational graphs as supported by deep learning engines, CUDA implementations or multi-threaded CPU processes. We show that MSRL subsumes the distribution strategies of existing systems, while scaling RL training to 64 GPUs.

cs.LG↗

arXiv2022

Extended Dynamical Mean Field Theory for Correlated Electron Models

An overarching question in strongly correlated electron systems is how the landscape of quantum phases emerges from electron correlations. The method of extended dynamical mean field theory (EDMFT) has been developed for clean lattice models of the correlated electrons. For such models, not only onsite Hubbard-like interactions are important, but so are intersite interactions. Importantly, the EDMFT method treats the interplay between the onsite and intersite interactions dynamically. It was initially formulated for models of the two-band Anderson-lattice type with intersite interactions, as well as for the one-band Hubbard type with intersite Heisenberg-like terms that are often called Hubbard-Heisenberg models. In the case of Kondo lattice models, the EDMFT method incorporates a dynamical competition between the local Kondo and intersite Ruderman-Kittel-Kasuya-Yosida (RKKY) interactions. In these models, the EDMFT-based analyses led to the notion of Kondo destruction, which has played a central role in the understanding of quantum critical heavy fermion metals. In this article, we summarize the EDMFT method, and survey its applications, particularly for Kondo/Anderson lattice models. We also discuss the prospect for further developing the EDMFT method, as well as for applying it to address the correlation physics in a variety of new settings. Among the latter are the orbital-selective Mott physics that arises both in iron-based superconductors and in frustrated bulk systems with topological flat bands.

cond-mat.str-el↗

arXiv2022

From braid groups to mapping class groups

In this paper, we classify homomorphisms from the braid group of $n$ strands to the mapping class group of a genus $g$ surface. In particular, we show that when $g<n-2$, all representations are either cyclic or standard. Our result is sharp in the sense that when $g=n-2$, a generalization of the hyperelliptic representation appears, which is not cyclic or standard. This gives a classification of surface bundles over the configuration space of the complex plane. As a corollary, we partially recover the result of Aramayona-Souto, which classifies homomorphisms between mapping class groups, with a slight improvement.

math.GT↗

arXiv2022

A Comprehensive Survey on Trustworthy Recommender Systems

As one of the most successful AI-powered applications, recommender systems aim to help people make appropriate decisions in an effective and efficient way, by providing personalized suggestions in many aspects of our lives, especially for various human-oriented online services such as e-commerce platforms and social media sites. In the past few decades, the rapid developments of recommender systems have significantly benefited human by creating economic value, saving time and effort, and promoting social good. However, recent studies have found that data-driven recommender systems can pose serious threats to users and society, such as spreading fake news to manipulate public opinion in social media sites, amplifying unfairness toward under-represented groups or individuals in job matching services, or inferring privacy information from recommendation results. Therefore, systems' trustworthiness has been attracting increasing attention from various aspects for mitigating negative impacts caused by recommender systems, so as to enhance the public's trust towards recommender systems techniques. In this survey, we provide a comprehensive overview of Trustworthy Recommender systems (TRec) with a specific focus on six of the most important aspects; namely, Safety & Robustness, Nondiscrimination & Fairness, Explainability, Privacy, Environmental Well-being, and Accountability & Auditability. For each aspect, we summarize the recent related technologies and discuss potential research directions to help achieve trustworthy recommender systems in the future.

cs.IR↗