arXiv subjects

Chuan Chen

Publications and source records attributed to Chuan Chen.

72 records · Page 4Linked to original sources

arXiv2022

Distributed Evolution Strategies for Black-box Stochastic Optimization

This work concerns the evolutionary approaches to distributed stochastic black-box optimization, in which each worker can individually solve an approximation of the problem with nature-inspired algorithms. We propose a distributed evolution strategy (DES) algorithm grounded on a proper modification to evolution strategies, a family of classic evolutionary algorithms, as well as a careful combination with existing distributed frameworks. On smooth and nonconvex landscapes, DES has a convergence rate competitive to existing zeroth-order methods, and can exploit the sparsity, if applicable, to match the rate of first-order methods. The DES method uses a Gaussian probability model to guide the search and avoids the numerical issue resulted from finite-difference techniques in existing zeroth-order methods. The DES method is also fully adaptive to the problem landscape, as its convergence is guaranteed with any parameter setting. We further propose two alternative sampling schemes which significantly improve the sampling efficiency while leading to similar performance. Simulation studies on several machine learning problems suggest that the proposed methods show much promise in reducing the convergence time and improving the robustness to parameter settings.

cs.NE↗

arXiv2021

FedGL: Federated Graph Learning Framework with Global Self-Supervision

Graph data are ubiquitous in the real world. Graph learning (GL) tries to mine and analyze graph data so that valuable information can be discovered. Existing GL methods are designed for centralized scenarios. However, in practical scenarios, graph data are usually distributed in different organizations, i.e., the curse of isolated data islands. To address this problem, we incorporate federated learning into GL and propose a general Federated Graph Learning framework FedGL, which is capable of obtaining a high-quality global graph model while protecting data privacy by discovering the global self-supervision information during the federated training. Concretely, we propose to upload the prediction results and node embeddings to the server for discovering the global pseudo label and global pseudo graph, which are distributed to each client to enrich the training labels and complement the graph structure respectively, thereby improving the quality of each local model. Moreover, the global self-supervision enables the information of each client to flow and share in a privacy-preserving manner, thus alleviating the heterogeneity and utilizing the complementarity of graph data among different clients. Finally, experimental results show that FedGL significantly outperforms baselines on four widely used graph datasets.

cs.LG↗

arXiv2021

Tensor Completion via Convolutional Sparse Coding Regularization

Tensor data often suffer from missing value problem due to the complex high-dimensional structure while acquiring them. To complete the missing information, lots of Low-Rank Tensor Completion (LRTC) methods have been proposed, most of which depend on the low-rank property of tensor data. In this way, the low-rank component of the original data could be recovered roughly. However, the shortcoming is that the detail information can not be fully restored, no matter the Sum of the Nuclear Norm (SNN) nor the Tensor Nuclear Norm (TNN) based methods. On the contrary, in the field of signal processing, Convolutional Sparse Coding (CSC) can provide a good representation of the high-frequency component of the image, which is generally associated with the detail component of the data. Nevertheless, CSC can not handle the low-frequency component well. To this end, we propose two novel methods, LRTC-CSC-I and LRTC-CSC-II, which adopt CSC as a supplementary regularization for LRTC to capture the high-frequency components. Therefore, the LRTC-CSC methods can not only solve the missing value problem but also recover the details. Moreover, the regularizer CSC can be trained with small samples due to the sparsity characteristic. Extensive experiments show the effectiveness of LRTC-CSC methods, and quantitative evaluation indicates that the performance of our models are superior to state-of-the-art methods.

cs.CV↗

arXiv2021

Competition of Spinon Fermi Surface and Heavy Fermi Liquids states from the Periodic Anderson to the Hubbard model

We study a model of correlated electrons coupled by tunnelling to a layer of itinerant metallic electrons, which allows to interpolate from a frustrated limit favorable to spin liquid states to a Kondo-lattice limit favorable to interlayer coherent heavy metallic states. We study the competition of the spinon fermi surface state and the interlayer coherent heavy Kondo metal that appears with increasing tunnelling. Employing a slave rotor mean-field approach, we obtain a phase diagram and describe two regimes where the spin liquid state is destroyed by weak interlayer tunnelling, (i) the Kondo limit in which the correlated electrons can be viewed as localized spin moments and (ii) near the Mott metal-insulator-transition where the spinon Fermi surface transitions continuously into a Fermi liquid. We study the shape of LDOS spectra of the putative spin liquid layer in the heavy Fermi liquid phase and describe the temperature dependence of its width arising from quasiparticle interactions and disorder effects throughout this phase diagram, in an effort to understand recent STM experiments of the candidate spin liquid 1T-TaSe$_2$ residing on metallic 1H-TaSe$_2$. Comparison of the shape and temperature dependence of the theoretical and experimental LDOS suggest that this system is either close to the localized Kondo limit, or in an intermediate coupling regime where the Kondo coupling and the Heisenberg exchange interaction are comparable.

cond-mat.str-el↗

arXiv2021

Outlier-Resilient Web Service QoS Prediction

The proliferation of Web services makes it difficult for users to select the most appropriate one among numerous functionally identical or similar service candidates. Quality-of-Service (QoS) describes the non-functional characteristics of Web services, and it has become the key differentiator for service selection. However, users cannot invoke all Web services to obtain the corresponding QoS values due to high time cost and huge resource overhead. Thus, it is essential to predict unknown QoS values. Although various QoS prediction methods have been proposed, few of them have taken outliers into consideration, which may dramatically degrade the prediction performance. To overcome this limitation, we propose an outlier-resilient QoS prediction method in this paper. Our method utilizes Cauchy loss to measure the discrepancy between the observed QoS values and the predicted ones. Owing to the robustness of Cauchy loss, our method is resilient to outliers. We further extend our method to provide time-aware QoS prediction results by taking the temporal information into consideration. Finally, we conduct extensive experiments on both static and dynamic datasets. The results demonstrate that our method is able to achieve better performance than state-of-the-art baseline methods.

cs.IR↗

arXiv2020

Who Are the Phishers? Phishing Scam Detection on Ethereum via Network Embedding

Recently, blockchain technology has become a topic in the spotlight but also a hotbed of various cybercrimes. Among them, phishing scams on blockchain have been found making a notable amount of money, thus emerging as a serious threat to the trading security of the blockchain ecosystem. In order to create a favorable environment for investment, an effective method for detecting phishing scams is urgently needed in the blockchain ecosystem. To this end, this paper proposes an approach to detect phishing scams on Ethereum by mining its transaction records. Specifically, we first crawl the labeled phishing addresses from two authorized websites and reconstruct the transaction network according to the collected transaction records. Then, by taking the transaction amount and timestamp into consideration, we propose a novel network embedding algorithm called trans2vec to extract the features of the addresses for subsequent phishing identification. Finally, we adopt the oneclass support vector machine (SVM) to classify the nodes into normal and phishing ones. Experimental results demonstrate that the phishing detection method works effectively on Ethereum, and indicate the efficacy of trans2vec over existing state-of-the-art algorithms on feature extraction for transaction networks. This work is the first investigation on phishing detection on Ethereum via network embedding and provides insights into how features of large-scale transaction networks can be embedded.

cs.SI↗

arXiv2020

Anatomy of Z2 fluxes in anyon Fermi liquids and Bose condensates

We study in detail the properties of pi-fluxes embedded in a state with a finite density of anyons that form either a Fermi liquid or a Bose-Einstein condensate. By employing a recently developed exact lattice bosonization in 2D, we demonstrate that such pi-flux remains a fully deconfined quasiparticle with a finite energy cost in a Fermi liquid of emergent fermions coupled to a Z2 gauge field. This pi-flux is accompanied by a screening cloud of fermions, which in the case of a Fermi gas with a parabolic dispersion binds exactly 1/8 of a fermionic hole. In addition there is a long-ranged power-law oscillatory disturbance of the liquid surrounding the pi-flux akin to Friedel oscillations. These results carry over directly to the pi-flux excitations in orthogonal metals. In sharp contrast, when the pi-flux is surrounded by a Bose-Einstein condensate of particles coupled to a Z2 gauge field, it binds a superfluid half-vortex, becoming a marginally confined excitation with a logarithmic energy cost divergence.

cond-mat.str-el↗

arXiv2020

An Uncoupled Training Architecture for Large Graph Learning

Graph Convolutional Network (GCN) has been widely used in graph learning tasks. However, GCN-based models (GCNs) is an inherently coupled training framework repetitively conducting the complex neighboring aggregation, which leads to the limitation of flexibility in processing large-scale graph. With the depth of layers increases, the computational and memory cost of GCNs grow explosively due to the recursive neighborhood expansion. To tackle these issues, we present Node2Grids, a flexible uncoupled training framework that leverages the independent mapped data for obtaining the embedding. Instead of directly processing the coupled nodes as GCNs, Node2Grids supports a more efficacious method in practice, mapping the coupled graph data into the independent grid-like data which can be fed into the efficient Convolutional Neural Network (CNN). This simple but valid strategy significantly saves memory and computational resource while achieving comparable results with the leading GCN-based models. Specifically, by ranking each node's influence through degree, Node2Grids selects the most influential first-order as well as second-order neighbors with central node fusion information to construct the grid-like data. For further improving the efficiency of downstream tasks, a simple CNN-based neural network is employed to capture the significant information from the mapped grid-like data. Moreover, the grid-level attention mechanism is implemented, which enables implicitly specifying the different weights for neighboring nodes with different influences. In addition to the typical transductive and inductive learning tasks, we also verify our framework on million-scale graphs to demonstrate the superiority of the proposed Node2Grids model against the state-of-the-art GCN-based approaches.

cs.LG↗

arXiv2020

SocialTrans: A Deep Sequential Model with Social Information for Web-Scale Recommendation Systems

On social network platforms, a user's behavior is based on his/her personal interests, or influenced by his/her friends. In the literature, it is common to model either users' personal preference or their socially influenced preference. In this paper, we present a novel deep learning model SocialTrans for social recommendations to integrate these two types of preferences. SocialTrans is composed of three modules. The first module is based on a multi-layer Transformer to model users' personal preference. The second module is a multi-layer graph attention neural network (GAT), which is used to model the social influence strengths between friends in social networks. The last module merges users' personal preference and socially influenced preference to produce recommendations. Our model can efficiently fit large-scale data and we deployed SocialTrans to a major article recommendation system in China. Experiments on three data sets verify the effectiveness of our model and show that it outperforms state-of-the-art social recommendation methods.

cs.IR↗

arXiv2020

Correlated states of a triangular net of coupled quantum wires: Implications for the phase diagram of marginally twisted bilayer graphene

We explore in detail the electronic phases of a system consisting of three non-colinear arrays of coupled quantum wires, each rotated 120 degrees with respect to the next. A perturbative renormalization-group analysis reveals that multiple correlated states can be stabilized: a $s$-wave or $d \pm id$ superconductor, a charge density wave insulator, a two-dimensional Fermi liquid, and a 2D Luttinger liquid (also known as smectic metal or sliding Luttinger liquid). The model provides an effective description of electronic interactions in small-angle twisted bilayer graphene and we discuss its implications in relation to the recent observation of correlated and superconducting groundstates near commensurate densities in magic-angle twisted samples, as well as the ``strange metal'' behavior at finite temperatures as a natural outcome of the 2D Luttinger liquid phase.

cond-mat.str-el↗

arXiv2020

A Blockchain-based Decentralized Federated Learning Framework with Committee Consensus

Federated learning has been widely studied and applied to various scenarios. In mobile computing scenarios, federated learning protects users from exposing their private data, while cooperatively training the global model for a variety of real-world applications. However, the security of federated learning is increasingly being questioned, due to the malicious clients or central servers' constant attack to the global model or user privacy data. To address these security issues, we proposed a decentralized federated learning framework based on blockchain, i.e., a Blockchain-based Federated Learning framework with Committee consensus (BFLC). The framework uses blockchain for the global model storage and the local model update exchange. To enable the proposed BFLC, we also devised an innovative committee consensus mechanism, which can effectively reduce the amount of consensus computing and reduce malicious attacks. We then discussed the scalability of BFLC, including theoretical security, storage optimization, and incentives. Finally, we performed experiments using real-world datasets to verify the effectiveness of the BFLC framework.

cs.DC↗

arXiv2020

LoCEC: Local Community-based Edge Classification in Large Online Social Networks

Relationships in online social networks often imply social connections in the real world. An accurate understanding of relationship types benefits many applications, e.g. social advertising and recommendation. Some recent attempts have been proposed to classify user relationships into predefined types with the help of pre-labeled relationships or abundant interaction features on relationships. Unfortunately, both relationship feature data and label data are very sparse in real social platforms like WeChat, rendering existing methods inapplicable. In this paper, we present an in-depth analysis of WeChat relationships to identify the major challenges for the relationship classification task. To tackle the challenges, we propose a Local Community-based Edge Classification (LoCEC) framework that classifies user relationships in a social network into real-world social connection types. LoCEC enforces a three-phase processing, namely local community detection, community classification and relationship classification, to address the sparsity issue of relationship features and relationship labels. Moreover, LoCEC is designed to handle large-scale networks by allowing parallel and distributed processing. We conduct extensive experiments on the real-world WeChat network with hundreds of billions of edges to validate the effectiveness and efficiency of LoCEC.

cs.SI↗

arXiv2019

Anomalous quantum metal in a 2D crystalline superconductor with intrinsic electronic non-uniformity

The details of the superconducting to quantum metal transition (SQMT) at T=0 are an open problem that invokes much interest in the nature of this exotic and unexpected ground state1-3. However, the SQMT was not yet investigated in a crystalline 2D superconductor with coexisting and fluctuating quantum orders. Here, we report the observation of a SQMT in 2D ion-gel gated 1T-TiSe24, driven by magnetic field. A field-induced crossover between Bose quantum metal and vortex quantum creeping with increasing field is observed. We discuss the interplay between superconducting and CDW fluctuations (discommensurations) and their relation to the anomalous quantum metal (AQM) phase. From our findings, gate-tunable 1T-TiSe2 emerges as a privileged platform to scrutinize, in a controlled way, the details of the SQMT, the role of coexisting fluctuating orders and, ultimately, obtain a deeper understanding of the fate of superconductivity in strictly two-dimensional crystals near zero temperature.

cond-mat.supr-con↗

arXiv2019

Discommensuration-enhanced superconductivity in the charge density wave phases of transition-metal dichalcogenides

We introduce a McMillan-Ginzburg-Landau theory to describe the cooperative coexistence of charge-density and superconducting order in two-dimensional crystals. With a free-energy that explicitly accounts for the competition between commensurate and incommensurate ground states, we are able to map the transition between these phases and monitor the development of discommensurations in the near-commensurate regime. Attributing the enhancement of superconducting order to density-wave fluctuations, we propose a coupling scheme that yields a phase diagram in qualitative agreement with experiments in conducting transition metal dichalcogenides. The model predicts the development of non-uniform superconductivity similar to that arising from a pair-density wave, with a spatial texture driven by the underlying charge-density wave fluctuations.

cond-mat.mes-hall↗

arXiv2019

Antiferromagnetism and chiral d-wave superconductivity from an effective $t-J-D$ model for twisted bilayer graphene

Starting from the strong-coupling limit of an extended Hubbard model, we develop a spin-fermion theory to study the insulating phase and pairing symmetry of the superconducting phase in twisted bilayer graphene. Assuming that the insulating phase is an anti-ferromagnetic insulator, we show that fluctuations of the anti-ferromagnetic order in the conducting phase can mediate superconducting pairing. Using a self-consistent mean-field analysis, we find that the pairing wave function has a chiral d-wave symmetry. Consistent with this observation, we show explicitly the existence of chiral Majorana edge modes by diagonalizing our proposed Hamiltonian on a finite-sized system. These results establish twisted bilayer graphene as a promising platform to realize topological superconductivity.

cond-mat.supr-con↗

arXiv2019

Market Manipulation of Bitcoin: Evidence from Mining the Mt. Gox Transaction Network

The cryptocurrency market is a very huge market without effective supervision. It is of great importance for investors and regulators to recognize whether there are market manipulation and its manipulation patterns. This paper proposes an approach to mine the transaction networks of exchanges for answering this question.By taking the leaked transaction history of Mt. Gox Bitcoin exchange as a sample,we first divide the accounts into three categories according to its characteristic and then construct the transaction history into three graphs. Many observations and findings are obtained via analyzing the constructed graphs. To evaluate the influence of the accounts' transaction behavior on the Bitcoin exchange price,the graphs are reconstructed into series and reshaped as matrices. By using singular value decomposition (SVD) on the matrices, we identify many base networks which have a great correlation with the price fluctuation. When further analyzing the most important accounts in the base networks, plenty of market manipulation patterns are found. According to these findings, we conclude that there was serious market manipulation in Mt. Gox exchange and the cryptocurrency market must strengthen the supervision.

q-fin.GN↗

arXiv2018

Reproduction of the charge density wave phase diagram in $1T$-$\mathrm{TiSe}_2$ exposes its excitonic character

Recent experiments suggest that excitonic degrees of freedom play an important role in precipitating the charge density wave (CDW) transition in $1T$-$\mathrm{TiSe}_2$. Through systematic calculations of the electronic and phonon spectrum based on density functional perturbation theory, we show that the predicted critical doping of the CDW phase overshoots the experimental value by 1 order of magnitude. In contrast, an independent self-consistent many-body calculation of the excitonic order parameter and renormalized band structure is able to capture the experimental phase diagram in extremely good qualitative and quantitative agreement. This demonstrates that electron-electron interactions and the excitonic instability arising from direct electron-hole coupling are pivotal to accurately describe the nature of the CDW in this system. This has important implications to understand the emergence of superconductivity within the CDW phase of this and related systems.

cond-mat.mes-hall↗

arXiv2018

Collaborative Deep Learning Across Multiple Data Centers

Valuable training data is often owned by independent organizations and located in multiple data centers. Most deep learning approaches require to centralize the multi-datacenter data for performance purpose. In practice, however, it is often infeasible to transfer all data to a centralized data center due to not only bandwidth limitation but also the constraints of privacy regulations. Model averaging is a conventional choice for data parallelized training, but its ineffectiveness is claimed by previous studies as deep neural networks are often non-convex. In this paper, we argue that model averaging can be effective in the decentralized environment by using two strategies, namely, the cyclical learning rate and the increased number of epochs for local model training. With the two strategies, we show that model averaging can provide competitive performance in the decentralized mode compared to the data-centralized one. In a practical environment with multiple data centers, we conduct extensive experiments using state-of-the-art deep network architectures on different types of data. Results demonstrate the effectiveness and robustness of the proposed method.

cs.LG↗