arXiv subjects

Sheng Zhang

Publications and source records attributed to Sheng Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2022

Client Selection and Bandwidth Allocation for Federated Learning: An Online Optimization Perspective

Federated learning (FL) can train a global model from clients' local data set, which can make full use of the computing resources of clients and performs more extensive and efficient machine learning on clients with protecting user information requirements. Many existing works have focused on optimizing FL accuracy within the resource constrained in each individual round, however there are few works comprehensively consider the optimization for latency, accuracy and energy consumption over all rounds in wireless federated learning. Inspired by this, in this paper, we investigate FL in wireless network where client selection and bandwidth allocation are two crucial factors which significantly affect the latency, accuracy and energy consumption of clients. We formulate the optimization problem as a mixed-integer problem, which is to minimize the cost of time and accuracy within the long-term energy constrained over all rounds. To address this optimization, we propose the Perround Energy Drift Plus Cost (PEDPC) algorithm in an online perspective, and the performance of the PEDPC algorithm is verified in simulation results in terms of latency, accuracy and energy consumption in IID and NON-IID dat distributions.

cs.NI↗

arXiv2022

2-d signature of images and texture classification

We introduce a proper notion of 2-dimensional signature for images. This object is inspired by the so-called rough paths theory, and it captures many essential features of a 2-dimensional object such as an image. It thus serves as a low-dimensional feature for pattern classification. Here we implement a simple procedure for texture classification. In this context, we show that a low dimensional set of features based on signatures produces an excellent accuracy.

cs.CV↗

arXiv2022

A Multi-User Effective Computation Offloading Mechanism for MEC System: Batched Multi-Armed Bandits Approach

With the development of 5G technology, mobile edge computing (MEC) is becoming a useful architecture, which is envisioned as a cloud computing extension version. Users within MEC system could deal with data processing at edge terminals, which can reduce time for communication or data transmission. Multi-armed bandits (MAB) algorithms are powerful tools helping users offloading tasks to their best servers in MEC. However, as the number of users and tasks growing, the frequency of selecting servers and the cost of making decision is growing rapidly under traditional MAB algorithms. Inspired by this, in this paper, we propose a Batch-based Multi-user Server Elimination (BMSE) algorithm to solve such problem, which includes two sub-algorithms. We firstly propose a sub-algorithm in user level (BMSE-UL) to reduce the time cost. In BMSE-UL, users can simplify its own available server groups and offload tasks collectively. Then another sub-algorithm in system level (BMSE-SL) is proposed to reduce the frequency of making decision. In BMSE-SL, the system can cut down all the suboptimal task offloading actions and make the choosing option unique. Furthermore, we establish the optimality of the proposed algorithms by proving the sub-linearity convergence of their regrets and demonstrate the effectiveness of BMSE by extensive experiments.

cs.NI↗

arXiv2022

A Review on Graph Neural Network Methods in Financial Applications

With multiple components and relations, financial data are often presented as graph data, since it could represent both the individual features and the complicated relations. Due to the complexity and volatility of the financial market, the graph constructed on the financial data is often heterogeneous or time-varying, which imposes challenges on modeling technology. Among the graph modeling technologies, graph neural network (GNN) models are able to handle the complex graph structure and achieve great performance and thus could be used to solve financial tasks. In this work, we provide a comprehensive review of GNN models in recent financial context. We first categorize the commonly-used financial graphs and summarize the feature processing step for each node. Then we summarize the GNN methodology for each graph type, application in each area, and propose some potential research areas.

q-fin.ST↗

arXiv2022

Locally Aggregated Feature Attribution on Natural Language Model Understanding

With the growing popularity of deep-learning models, model understanding becomes more important. Much effort has been devoted to demystify deep neural networks for better interpretability. Some feature attribution methods have shown promising results in computer vision, especially the gradient-based methods where effectively smoothing the gradients with reference data is key to a robust and faithful result. However, direct application of these gradient-based methods to NLP tasks is not trivial due to the fact that the input consists of discrete tokens and the "reference" tokens are not explicitly defined. In this work, we propose Locally Aggregated Feature Attribution (LAFA), a novel gradient-based feature attribution method for NLP models. Instead of relying on obscure reference tokens, it smooths gradients by aggregating similar reference texts derived from language model embeddings. For evaluation purpose, we also design experiments on different NLP tasks including Entity Recognition and Sentiment Analysis on public datasets as well as key feature detection on a constructed Amazon catalogue dataset. The superior performance of the proposed method is demonstrated through experiments.

cs.CL↗

arXiv2022

Machine learning predictions for local electronic properties of disordered correlated electron systems

We present a scalable machine learning (ML) model to predict local electronic properties such as on-site electron number and double occupation for disordered correlated electron systems. Our approach is based on the locality principle, or the nearsightedness nature, of many-electron systems, which means local electronic properties depend mainly on the immediate environment. A ML model is developed to encode this complex dependence of local quantities on the neighborhood. We demonstrate our approach using the square-lattice Anderson-Hubbard model, which is a paradigmatic system for studying the interplay between Mott transition and Anderson localization. We develop a lattice descriptor based on group-theoretical method to represent the on-site random potentials within a finite region. The resultant feature variables are used as input to a multi-layer fully connected neural network, which is trained from datasets of variational Monte Carlo (VMC) simulations on small systems. We show that the ML predictions agree reasonably well with the VMC data. Our work underscores the promising potential of ML methods for multi-scale modeling of correlated electron systems.

cond-mat.str-el↗

arXiv2022

Zero-Shot Dependency Parsing with Worst-Case Aware Automated Curriculum Learning

Large multilingual pretrained language models such as mBERT and XLM-RoBERTa have been found to be surprisingly effective for cross-lingual transfer of syntactic parsing models (Wu and Dredze 2019), but only between related languages. However, source and training languages are rarely related, when parsing truly low-resource languages. To close this gap, we adopt a method from multi-task learning, which relies on automated curriculum learning, to dynamically optimize for parsing performance on outlier languages. We show that this approach is significantly better than uniform and size-proportional sampling in the zero-shot setting.

cs.CL↗

arXiv2022

FairLex: A Multilingual Benchmark for Evaluating Fairness in Legal Text Processing

We present a benchmark suite of four datasets for evaluating the fairness of pre-trained language models and the techniques used to fine-tune them for downstream tasks. Our benchmarks cover four jurisdictions (European Council, USA, Switzerland, and China), five languages (English, German, French, Italian and Chinese) and fairness across five attributes (gender, age, region, language, and legal area). In our experiments, we evaluate pre-trained language models using several group-robust fine-tuning techniques and show that performance group disparities are vibrant in many cases, while none of these techniques guarantee fairness, nor consistently mitigate group disparities. Furthermore, we provide a quantitative and qualitative analysis of our results, highlighting open challenges in the development of robustness methods in legal NLP.

cs.CL↗

arXiv2022

Quantum-Memory-Enhanced Preparation of Nonlocal Graph States

Graph states are an important class of multipartite entangled states. Previous experimental generation of graph states and in particular the Greenberger-Horne-Zeilinger (GHZ) states in linear optics quantum information schemes is subjected to an exponential decay in efficiency versus the system size, which limits its large-scale applications in quantum networks. Here we demonstrate an efficient scheme to prepare graph states with only a polynomial overhead using long-lived atomic quantum memories. We generate atom-photon entangled states in two atomic ensembles asynchronously, retrieve the stored atomic excitations only when both sides succeed, and further project them into a four-photon GHZ state. We measure the fidelity of this GHZ state and further demonstrate its applications in the violation of Bell-type inequalities and in quantum cryptography. Our work demonstrates the prospect of efficient generation of multipartite entangled states in large-scale distributed systems with applications in quantum information processing and metrology.

quant-ph↗

arXiv2022

Rule Mining over Knowledge Graphs via Reinforcement Learning

Knowledge graphs (KGs) are an important source repository for a wide range of applications and rule mining from KGs recently attracts wide research interest in the KG-related research community. Many solutions have been proposed for the rule mining from large-scale KGs, which however are limited in the inefficiency of rule generation and ineffectiveness of rule evaluation. To solve these problems, in this paper we propose a generation-then-evaluation rule mining approach guided by reinforcement learning. Specifically, a two-phased framework is designed. The first phase aims to train a reinforcement learning agent for rule generation from KGs, and the second is to utilize the value function of the agent to guide the step-by-step rule generation. We conduct extensive experiments on several datasets and the results prove that our rule mining solution achieves state-of-the-art performance in terms of efficiency and effectiveness.

cs.AI↗

arXiv2022

Experimental demonstration of memory-enhanced scaling for entanglement connection of quantum repeater segments

The quantum repeater protocol is a promising approach to implement long-distance quantum communication and large-scale quantum networks. A key idea of the quantum repeater protocol is to use long-lived quantum memories to achieve efficient entanglement connection between different repeater segments with a polynomial scaling. Here we report an experiment which realizes efficient connection of two quantum repeater segments via on-demand entanglement swapping by the use of two atomic quantum memories with storage time of tens of milliseconds. With the memory enhancement, scaling-changing acceleration is demonstrated in the rate for a successful entanglement connection. The experimental realization of entanglement connection of two quantum repeater segments with an efficient memory-enhanced scaling demonstrates a key advantage of the quantum repeater protocol, which makes a cornerstone towards future large-scale quantum networks.

quant-ph↗

arXiv2022

Finite-Sample Analysis of Nonlinear Stochastic Approximation with Applications in Reinforcement Learning

Motivated by applications in reinforcement learning (RL), we study a nonlinear stochastic approximation (SA) algorithm under Markovian noise, and establish its finite-sample convergence bounds under various stepsizes. Specifically, we show that when using constant stepsize (i.e., $α_k\equiv α$), the algorithm achieves exponential fast convergence to a neighborhood (with radius $O(α\log(1/α))$) around the desired limit point. When using diminishing stepsizes with appropriate decay rate, the algorithm converges with rate $O(\log(k)/k)$. Our proof is based on Lyapunov drift arguments, and to handle the Markovian noise, we exploit the fast mixing of the underlying Markov chain. To demonstrate the generality of our theoretical results on Markovian SA, we use it to derive the finite-sample bounds of the popular $Q$-learning with linear function approximation algorithm, under a condition on the behavior policy. Importantly, we do not need to make the assumption that the samples are i.i.d., and do not require an artificial projection step in the algorithm to maintain the boundedness of the iterates. Numerical simulations corroborate our theoretical results.

math.OC↗

arXiv2022

Can depth-adaptive BERT perform better on binary classification tasks

In light of the success of transferring language models into NLP tasks, we ask whether the full BERT model is always the best and does it exist a simple but effective method to find the winning ticket in state-of-the-art deep neural networks without complex calculations. We construct a series of BERT-based models with different size and compare their predictions on 8 binary classification tasks. The results show there truly exist smaller sub-networks performing better than the full model. Then we present a further study and propose a simple method to shrink BERT appropriately before fine-tuning. Some extended experiments indicate that our method could save time and storage overhead extraordinarily with little even no accuracy loss.

cs.CL↗

arXiv2022

Descriptors for Machine Learning Model of Generalized Force Field in Condensed Matter Systems

We outline the general framework of machine learning (ML) methods for multi-scale dynamical modeling of condensed matter systems, and in particular of strongly correlated electron models. Complex spatial temporal behaviors in these systems often arise from the interplay between quasi-particles and the emergent dynamical classical degrees of freedom, such as local lattice distortions, spins, and order-parameters. Central to the proposed framework is the ML energy model that, by successfully emulating the time-consuming electronic structure calculation, can accurately predict a local energy based on the classical field in the intermediate neighborhood. In order to properly include the symmetry of the electron Hamiltonian, a crucial component of the ML energy model is the descriptor that transforms the neighborhood configuration into invariant feature variables, which are input to the learning model. A general theory of the descriptor for the classical fields is formulated, and two types of models are distinguished depending on the presence or absence of an internal symmetry for the classical field. Several specific approaches to the descriptor of the classical fields are presented. Our focus is on the group-theoretical method that offers a systematic and rigorous approach to compute invariants based on the bispectrum coefficients. We propose an efficient implementation of the bispectrum method based on the concept of reference irreducible representations. Finally, the implementations of the various descriptors are demonstrated on well-known electronic lattice models.

cond-mat.str-el↗

arXiv2021

DVHN: A Deep Hashing Framework for Large-scale Vehicle Re-identification

In this paper, we make the very first attempt to investigate the integration of deep hash learning with vehicle re-identification. We propose a deep hash-based vehicle re-identification framework, dubbed DVHN, which substantially reduces memory usage and promotes retrieval efficiency while reserving nearest neighbor search accuracy. Concretely,~DVHN directly learns discrete compact binary hash codes for each image by jointly optimizing the feature learning network and the hash code generating module. Specifically, we directly constrain the output from the convolutional neural network to be discrete binary codes and ensure the learned binary codes are optimal for classification. To optimize the deep discrete hashing framework, we further propose an alternating minimization method for learning binary similarity-preserved hashing codes. Extensive experiments on two widely-studied vehicle re-identification datasets- \textbf{VehicleID} and \textbf{VeRi}-~have demonstrated the superiority of our method against the state-of-the-art deep hash methods. \textbf{DVHN} of $2048$ bits can achieve 13.94\% and 10.21\% accuracy improvement in terms of \textbf{mAP} and \textbf{Rank@1} for \textbf{VehicleID (800)} dataset. For \textbf{VeRi}, we achieve 35.45\% and 32.72\% performance gains for \textbf{Rank@1} and \textbf{mAP}, respectively.

cs.CV↗

arXiv2021

Anomalous phase separation dynamics in a correlated electron system: machine-learning enabled large-scale kinetic Monte Carlo simulations

Phase separation plays a central role in the emergence of novel functionalities of correlated electron materials. The structure of the mixed-phase states depends strongly on the nonequilibrium phase-separation dynamics, which has so far yet to be systematically investigated, especially on the theoretical side. With the aid of modern machine learning methods, we demonstrate the first-ever large-scale kinetic Monte Carlo simulations of the phase separation process for the Falicov-Kimball model, which is one of the canonical strongly correlated electron systems. We uncover an unusual phase-separation scenario where domain coarsening occurs simultaneously at two different scales: the growth of checkerboard clusters at smaller length scales and the expansion of super-clusters, which are aggregates of the checkerboard clusters of the same sign, at a larger scale. We show that the emergence of super-clusters is due to a hidden dynamical breaking of the sublattice symmetry. Arrested growth of the checkerboard patterns and of the super-clusters is shown to result from a correlation-induced self-trapping mechanism. Glassy behaviors similar to the one reported in this work could be generic for other correlated electron systems.

cond-mat.str-el↗

arXiv2021

An unconstrained-like control-based dynamic method for optimization problems with simple bounds

The optimization problems with simple bounds are an important class of problems. To facilitate the computation of such problems, an unconstrained-like dynamic method, motivated by the Lyapunov control principle, is proposed. This method employs the anti-windup limited integrator to address the bounds of parameters upon the dynamics for unconstrained problem, and then solves the transformed Initial-value Problems (IVPs) with mature Ordinary Differential Equation (ODE) integration methods. It is proved that when the gain matrix is diagonal, the result is equivalent to that of the general dynamic method which involves an intermediate Quadratic Programming (QP) sub-problem. Thus, the global convergence to the optimal solution is guaranteed without the requirement of the strict complementarity condition. Since the estimation of the right active constraints is avoided and no programming sub-problem is involved in the computation process, it shows higher efficiency than the general dynamic method and other common iterative methods through the numerical examples. In particular, the implementation is simple, and the proposed method is easy-to-use.

math.OC↗

arXiv2021

Multicell Atomic Quantum Memory as a Hardware-Efficient Quantum Repeater Node

For scalable quantum communication and networks, a key step is to realize a quantum repeater node that can efficiently connect different segments of atom-photon entanglement using quantum memories. We report a compact and hardware-efficient realization of a quantum repeater node using a single atomic ensemble for multicell quantum memories. Millisecond lifetime is achieved for individual memory cells after suppressing the magnetic-field-induced inhomogeneous broadening and the atomic-motion-induced spin-wave dephasing. Based on these long-lived multicell memory cells, we achieve heralded asynchronous entanglement generation in two quantum repeater segments one after another and then an on-demand entanglement connection of these two repeater segments. As another application of the multicell atomic quantum memory, we further demonstrate storage and on-demand retrieval of heralded atomic spin-wave qubits by implementing a random access quantum memory with individual addressing capacity. This work provides a promising constituent for efficient realization of quantum repeaters for large-scale quantum networks.

quant-ph↗