检索结果-内蒙古大学图书馆

Proximal algorithms for Smoothed Online Convex Optimization With Predictions

IEEE TRANSACTIONS ON signal processing 2023年 71卷 3442-3457页

作者： Senapati, Spandan Shenai, Ashwin Rajawat, Ketan Indian Inst Technol Kanpur Kanpur 208016 Uttar Pradesh India Univ Southern Calif Dept Comp Sci Los Angeles CA 90089 USA Swiss Fed Inst Technol Dept Comp Sci D INFK CH-8092 Zurich Switzerland Indian Inst Technol Kanpur Dept Elect Engn Kanpur 208016 Uttar Pradesh India

We consider a smoothed online convex optimization (SOCO) problem with predictions, where the learner has access to a finite lookahead window of time-varying stage costs, but suffers a switching cost for changing its actions at each stage. Based on the Alternating Proximal Gradient Descent (APGD) framework, we develop Receding Horizon Alternating Proximal Descent (RHAPD) for proximable, non-smooth and strongly convex stage costs, and RHAPD-Smooth (RHAPD-S) for non-proximable, smooth and strongly convex stage costs. In addition to outperforming gradient descent-based algorithms, while maintaining a comparable runtime complexity, our proposed algorithms also allow us to solve a wider range of problems. We provide theoretical upper bounds on the dynamic regret achieved by the proposed algorithms, which decay exponentially with the length of the lookahead window. The performance of the presented algorithms is empirically demonstrated via numerical experiments on non-smooth regression, dynamic trajectory tracking, and economic power dispatch problems.

关键词： Costs Heuristic algorithms Switches signal processing algorithms Prediction algorithms Convex functions Trajectory Smoothed online convex optimization predictions alternating proximal gradient descent

来源：评论

学校读者我要写书评

暂无评论

Performance of Cell-Free Massive MIMO Under Imperfect Channel State Information and Reciprocity Calibration

引用

IEEE SYSTEMS JOURNAL 2023年第3期17卷 4383-4394页

作者： Cao, Yang Liang, Xianghu Wang, Xiaoni Li, Yuan Fu, Qiang Wang, Dongming You, Xiaohu Southeast Univ Natl Mobile Commun Res Lab Nanjing 210096 Peoples R China Pervas Commun Ctr Purple Mt Labs Nanjing 211100 Peoples R China ZTE Corp State Key Lab Mobile Network & Mobile Multimedia T Shenzhen 518055 Peoples R China

This article studies the downlink performance of a cell-free massive multiple-input-multiple-output (MIMO) system with imperfect channel estimation and reciprocity calibration. We present the statistical characteristics of the calibration error with the Argos algorithm and a relay-based calibration algorithm to obtain the calibration parameters of the whole network. Then, we derive approximate expressions of the downlink spectral efficiency for a cell-free massive MIMO system with imperfect channel state information and calibration error. The numerical results show the advantage of the proposed calibration algorithm and validate the accuracy of the theoretical expressions.

关键词： Calibration Downlink Radio frequency Precoding Uplink signal processing algorithms Transceivers Cell-free massive multiple-input-multiple-output (MIMO) reciprocity calibration spectral efficiency (SE)

来源：评论

学校读者我要写书评

暂无评论

Most Resource Efficient Matrix Vector Multiplication on FPGAs

引用

IEEE ACCESS 2023年 11卷 3881-3898页

作者： Lehnert, Alexander Holzinger, Philipp Pfenning, Simon Mueller, Ralf Reichenbach, Marc Brandenburg Univ Technol Cottbus Senftenberg Chair Comp Engn D-03046 Cottbus Germany Friedrich Alexander Univ Erlangen Nurnberg Chair Comp Architecture D-91058 Erlangen Germany Friedrich Alexander Univ Erlangen Nurnberg Inst Digital Commun D-91058 Erlangen Germany

Fast and resource-efficient inference in artificial neural networks (ANNs) is of utmost importance and drives many new developments in the area of new hardware architectures, e.g., by means of systolic arrays or algorithmic optimization such as pruning. In this paper, we present a novel method for lowering the computation effort for ANN inference utilizing ideas from information theory. Weight matrices are sliced into submatrices of logarithmic aspect ratios. These slices are then factorized. This reduces the number of required computations without compromising on fully parallel processing. We create a new hardware architecture for this dedicated purpose. We also provide a tool to map these sliced and factorized matrices efficiently to reconfigurable hardware. By comparing to the state of the art FPGA implementations, we can prove our claim by lowering hardware resources measured in look-up-tables (LUTs) by a factor of three to six. Our method does not rely on any particular property of the weight matrices of the ANN. It works for the general task of multiplying an input vector with a constant matrix and is also suitable for digital signal processing beyond ANNs.

关键词： Neural networks Computer architecture Matrix decomposition Field programmable gate arrays Sparse matrices signal processing algorithms Encoding Reconfigurable architectures Computational efficiency Constant matrix multiplication neural networks computer architecture reconfigurable architectures computational efficiency

来源：评论

学校读者我要写书评

暂无评论

On the Primal Feasibility in Dual Decomposition Methods Under Additive and Bounded Errors

引用

IEEE TRANSACTIONS ON signal processing 2023年 71卷 655-669页

作者： Abeynanda, Hansi Weeraddana, Chathuranga Lanel, G. H. J. Fischione, Carlo KTH Royal Inst Technol Sch Elect Engn & Comp Sci S-11428 Stockholm Sweden Univ Oulu Ctr Wireless Commun Oulu 90014 Finland Univ Sri Jayewardenepura Dept Math Nugegoda 10250 Sri Lanka KTH Royal Inst Technol Sch Elect Engn & Comp Sci & Digital Futures S-11428 Stockholm Sweden

With the unprecedented growth of signal processing and machine learning application domains, there has been a tremendous expansion of interest in distributed optimization methods to cope with the underlying large-scale problems. Nonetheless, inevitable system-specific challenges such as limited computational power, limited communication, latency requirements, measurement errors, and noises in wireless channels impose restrictions on the exactness of the underlying algorithms. Such restrictions have appealed to the exploration of algorithms' convergence behaviors under inexact settings. Despite the extensive research conducted in the area, it seems that the analysis of convergences of dual decomposition methods concerning primal optimality violations, together with dual optimality violations is less investigated. Here, we provide a systematic exposition of the convergence of feasible points in dual decomposition methods under inexact settings, for an important class of global consensus optimization problems. Convergences and the rate of convergences of the algorithms are mathematically substantiated, not only from a dual-domain standpoint but also from a primal-domain standpoint. Analytical results show that the algorithms converge to a neighborhood of optimality, the size of which depends on the level of underlying distortions.

关键词： Convergence signal processing algorithms Distortion Optimization Linear programming Wireless communication Quantization (signal) Distributed optimization inexact coordination federated learning

来源：评论

学校读者我要写书评

暂无评论

Communication-Efficient Federated Linear and Deep Generalized Canonical Correlation Analysis

引用

IEEE TRANSACTIONS ON signal processing 2023年 71卷 1379-1394页

作者： Shrestha, Sagar Fu, Xiao Oregon State Univ Sch Elect Engn & Comp Sci Corvallis OR 97331 USA

Classic and deep learning-based generalized canonical correlation analysis (GCCA) algorithms seek low-dimensional common representations of data entities from multiple "views" (e.g., audio and image) using linear transformations and neural networks, respectively. When the views are acquired and stored at different computing agents (e.g., organizations and edge devices) and data sharing is undesired due to privacy or communication cost considerations, federated learning-based GCCA is well-motivated. In federated learning, the views are kept locally at the agents and only derived, limited information exchange with a central server is allowed. However, applying existing GCCA algorithms onto such setting may incur prohibitively high communication overhead. This work puts forth a communication-efficient federated learning framework for both linear and deep GCCA under the maximum variance (MAX-VAR) formulation. The overhead issue is addressed by aggressively compressing (via quantization) the exchanging information between the computing agents and a central controller. Our synthetic and real-data experiments shows that the proposed algorithm enjoys a substantial reduction of communication overheads with virtually no loss in accuracy and convergence speed compared to the unquantized version. Rigorous convergence analyses are also presented, which is a nontrivial effort since generic federated optimization results do not cover the special problem structure of GCCA. Our result shows that the proposed algorithms for both linear and deep GCCA converge to critical points at a sublinear rate, even under heavy quantization and stochastic approximations. In addition, in the linear MAX-VAR case, the quantized algorithm approaches a global optimum in a geometric rate under reasonable conditions.

关键词： Convergence signal processing algorithms Optimization Federated learning Correlation Quantization (signal) Manifolds Generalized canonical correlation analysis communication-efficient federated learning nonconvex optimization convergence analysis

来源：评论

学校读者我要写书评

暂无评论

Generalized Unitary Approximate Message Passing for Double Linear Transformation Model

引用

IEEE TRANSACTIONS ON signal processing 2023年 71卷 1524-1538页

作者： Mo, Linlin Lu, Xinhua Yuan, Jide Zhang, Chuanzong Wang, Zhongyong Popovski, Petar Zhengzhou Univ Sch Elect & Informat Engn Zhengzhou 450001 Peoples R China Nanyang Inst Technol Coll Informat Engn Nanyang 473000 Peoples R China Soochow Univ Sch Elect & Informat Engn Suzhou 215006 Peoples R China Aalborg Univ Dept Elect Syst DK-9220 Aalborg Denmark

The double linear transformation model Y = AXB + W plays an important role in a variety of science and engineering applications, where X is estimated through known transformation matrices A and B from the noisy measurement Y. Decoupling X from Y is a formidable task due to the high complexity brought by the multiplication of the unknown matrix (vector) with the transformation matrix (M-UMTM). Unitary approximate message passing (UAMP) has been verified as a low complexity and strong robustness solution to the M-UMTM problems. However, it has only been used to tackle the problems with a single linear transformation matrix. In this work, we develop a generalized algorithm, namely, generalized double UAMP (GD-UAMP) for the target model, which not only inherits the low complexity of AMP, but also enhances robustness by employing double unitary transformation. As a generalized algorithm, GD-UAMP can be applied to address the generalized Bayesian inference problem, i.e., the arbitrary prior probability of X and likelihood function of Z, where Z = AXB is the noiseless measurement. We verify the feasibility of the proposed algorithm in the channel estimation problem for various wireless communication systems. Numerical results demonstrate that the proposed algorithm can perfectly fit different scenarios and showcase superior performance compared with benchmarks.

关键词： signal processing algorithms Approximation algorithms Message passing Complexity theory Inference algorithms Mathematical models Bayes methods Approximate message passing (AMP) channel estimation message passing Reconfigurable intelligent surface (RIS) unitary transformation

来源：评论

学校读者我要写书评

暂无评论

2-D Wavenumber Domain Autofocusing for High-Resolution Highly Squinted SAR Imaging Based on Equivalent Broadside Model

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2023年 61卷 1页

作者： Deng, Yuhui Sun, Guang-Cai Han, Liang Wang, Yuqi Zhang, Yu Xing, Mengdao Xidian Univ Natl Key Lab Radar Signal Proc Xian 710071 Peoples R China Xidian Univ Sch Optoelect Engn Xian 710071 Peoples R China Xian Univ Sci & Technol Sch Commun & Informat Engn Xian 710699 Peoples R China Xidian Univ Acad Adv Interdisciplinary Res Xian 710071 Peoples R China

The wavenumber domain algorithm is an ideal solution for high-resolution and highly squinted (HRHS) synthetic aperture radar (SAR) imaging in the case of an ideal straight trajectory. However, for airborne HRHS SAR imaging, the Stolt mapping leads to nonsystematic range cell migration (NsRCM) and secondary range compression (SRC) for the HRHS SAR, which causes the HRHS SAR image to defocus severely. To obtain a well-focused HRHS SAR image, a new autofocusing algorithm for HRHS SAR imagery based on an equivalent broadside model is proposed in this article. After coarse motion compensation, linear range walk correction and azimuth resampling are employed to transform the HRHS SAR data into the equivalent broadside SAR data, which has been proved to greatly reduce NsRCM and SRC. After that, the motion error prior structure in the 2-D wavenumber domain is revealed. According to the structure, a novel 2-D wavenumber domain autofocusing algorithm is proposed by the relationship between the 1-D azimuth phase error and the 2-D wavenumber domain phase error correction. Finally, the well-focused HRHS SAR imagery is obtained. Experiments based on simulated and acquired data are carried out to verify the necessity and effectiveness of the proposed algorithm for HRHS SAR imaging.

关键词： Radar polarimetry Azimuth Trajectory Radar imaging Radar Imaging signal processing algorithms 2-D wavenumber domain autofocusing azimuth resampling equivalent broadside model extended range migration algorithm (ERMA) high-resolution and highly squinted (HRHS) synthetic aperture radar (SAR)

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2023年第12期24卷 13841-13851页

作者： Shang, Wen-Long Tao, Xiaoming Bi, Huibo Chen, Yanyan Zhang, Hui Ochieng, Washington Y. Beijing Univ Technol Coll Metropolitan Transportat Beijing 100124 Peoples R China Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol Beijing 100084 Peoples R China China Acad Railway Sci Corp Ltd CARS Natl Railway Track Test Ctr Beijing 100081 Peoples R China Imperial Coll London Dept Civil & Environm Engn London SW7 2AZ England

The fast advancement of urban transportation systems in the recent decades has on one hand improved efficiency in traffic control and management, yet on the other hand brought new obstacles and interferences in audio related services in transportation systems, which is one of the dominating components in urban transportation systems, such as end-to-end Voice over Internet Protocol (VoIP) communications, risk alerting, and personalised recommendation services. The movement of vehicles/trains and the growing complexity of transportation infrastructures has become a big threat to the audio related services. Hence it is crucial to evaluate the Quality of Experience (QoE) of audio related services. Different from traditional algorithms which use digital signal processing to evaluate the QoE of mobile users, in this paper, we propose a two-stage brain-alike neural network aided graph learning algorithm to evaluate the QoE of audio signals with the aid of EEG feature extraction. The results are evaluated by newly-collected on-site data in public transportation environments and are examined by a branch of human experts to show that our algorithm outperforms other benchmark algorithms in term of human perception and accuracy of classification.

关键词： signal processing algorithms Transportation Quality of experience Electroencephalography Feature extraction Classification algorithms Deep learning Quality of experience evaluation environmental sound processing brain-inspired learning

来源：评论

学校读者我要写书评

暂无评论

A Provable Splitting Approach for Symmetric Nonnegative Matrix Factorization

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2023年第3期35卷 2206-2219页

作者： Li, Xiao Zhu, Zhihui Li, Qiuwei Liu, Kai Chinese Univ Hong Kong Sch Data Sci Shenzhen 518172 Guangdong Peoples R China Shenzhen Inst Artificial Intelligence & Robot Soc Shenzhen Guangdong Peoples R China Univ Denver Dept Elect & Comp Engn Denver CO 80208 USA Alibaba Grp Damo Acad Decis Intelligence Lab Bellevue WA 98004 USA Clemson Univ Comp Sci Div Clemson SC 29634 USA

The symmetric Nonnegative Matrix Factorization (NMF), a special but important class of the general NMF, has found numerous applications in data analysis such as various clustering tasks. Unfortunately, designing fast algorithms for the symmetric NMF is not as easy as for its nonsymmetric counterpart, since the latter admits the splitting property that allows state-of-the-art alternating-type algorithms. To overcome this issue, we first split the decision variable and transform the symmetric NMF to a penalized nonsymmetric one, paving the way for designing efficient alternating-type algorithms. We then show that solving the penalized nonsymmetric reformulation returns a solution to the original symmetric NMF. Moreover, we design a family of alternating-type algorithms and show that they all admit strong convergence guarantee: the generated sequence of iterates is convergent and converges at least sublinearly to a critical point of the original symmetric NMF. Finally, we conduct experiments on both synthetic data and real image clustering to support our theoretical results and demonstrate the performance of the alternating-type algorithms.

关键词： Convergence Symmetric matrices signal processing algorithms Clustering algorithms Minimization Transforms Optimization Symmetric nonnegative matrix factorization convergence image clustering alternating minimization

来源：评论

学校读者我要写书评

暂无评论

Distributed Banach-Picard Iteration: Application to Distributed Parameter Estimation and PCA

引用

IEEE TRANSACTIONS ON signal processing 2023年 71卷 17-30页

作者： Andrade, Francisco Figueiredo, Mario A. T. Xavier, Joao Univ Lisbon Inst Super Tecn P-1049001 Lisbon Portugal Inst Telecomunicacoes P-1049001 Lisbon Portugal Inst Syst & Robot Lab Robot & Engn Syst P-1600011 Lisbon Portugal

We recently proposed an algorithmic framework, distributed Banach-Picard iteration (DBPI), allowing a set of agents linked by a communication network to find a fixed point of a map that: (a) is the average of individual maps held by said agents;(b) is locally contractive (LC). Given such a map, DBPI yields a distributed algorithm provably inheriting the local linear convergence (LLC) of the standard Banach-Picard iteration for the centralized (average) map. Here, we instantiate DBPI in two classical problems, which amounts to proving that the conditions guaranteeing the LLC of DBPI hold. First, taking Sanger's algorithm for principal component analysis (PCA), we show that it corresponds to iterating an LC map that can be written as the average of local maps held by agents with private data subsets. Applying DBPI then recovers a previous distributed PCA algorithm, which lacked a convergence proof, thus closing that gap. In the second instantiation, we show that a variant of the expectation-maximization (EM) algorithm for parameter estimation from noisy, faulty measurements in sensor networks can be written as iterating an LC map that is the average of local maps. Consequently, the DBPI framework yields a distributed algorithm automatically inheriting the LLC guarantee of its centralized counterpart. Verifying the LC condition for EM is nontrivial (as the underlying operator depends on random samples) and a contribution in itself, possibly of independent interest. Finally, we illustrate experimentally the linear convergence of the proposed distributed EM algorithm, contrasting with the sub-linear rate of the previous version.

关键词： Task analysis Convergence Principal component analysis signal processing algorithms Distributed algorithms Dimensionality reduction Computational modeling Distributed computation banach-picard iteration fixed points distributed EM distributed PCA consensus

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：