检索结果-内蒙古大学图书馆

Projection-Free Stochastic Bi-Level Optimization

IEEE TRANSACTIONS ON signal processing 2022年 70卷 6332-6347页

作者： Akhtar, Zeeshan Bedi, Amrit Singh Thomdapu, Srujan Teja Rajawat, Ketan Indian Inst Technol Kanpur Dept Elect Engn Kanpur 208016 India Univ Maryland Inst Syst Res College Pk MD 20742 USA

Bi-level optimization, where the objective function depends on the solution of an inner optimization problem, provides a flexible framework for solving a rich class of problems such as hyper-parameter optimization and model-agnostic meta-learning. This work puts forth the first Stochastic Bi-level Frank-Wolfe (SBFW) algorithm to solve the stochastic bi-level optimization problems in a projection-free manner. We utilize a momentum based gradient tracker that results in a sample complexity of O(epsilon(-3)) for convex outer objectives and O(epsilon(-4)) for non-convex outer objectives with strongly convex inner objectives. The stochastic compositional optimization problems (a special case of bi-level optimization problems entailing the minimization of a composition of two expected-value functions) are also considered within the same rubric. The proposed Stochastic Compositional Frank-Wolfe (SCFW) algorithm is shown to achieve a sample complexity of O (epsilon(-2)) for convex objectives and O(epsilon(-3)) for non-convex objectives, at par with the state-of-the-art rates of projection-free algorithms for single-level problems. The usefulness and flexibility of SBFW and SCFW algorithms are demonstrated via extensive numerical tests. We show that SBFW outperforms the state-of-the-art methods for the problem of matrix completion with denoising, and achieve improvements of up to 82% in terms of the wall-clock time required to achieve a particular level of accuracy. Furthermore, we demonstrate the improved performance of SCFW over the competing projection-free variants on the policy evaluation problem in reinforcement learning.

关键词： Optimization signal processing algorithms Stochastic processes Noise reduction Complexity theory Noise measurement Data models Bi-level stochastic optimization frank-wolfe algorithm projection free algorithm

来源：评论

学校读者我要写书评

暂无评论

Joint Throughput-Power Optimization of Fog-RAN Using Rate-Splitting Multiple Access and Reinforcement-Learning Based User Clustering

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2021年第8期70卷 8019-8036页

作者： Hassan, Md. Zoheb Hossain, Md. Jahangir Cheng, Julian Leung, Victor C. M. Ecole Technol Super Montreal PQ H3C 1K3 Canada Univ British Columbia Sch Engn Kelowna BC V6T 1Z4 Canada Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen 518060 Peoples R China Univ British Columbia Dept Elect & Comp Engn Vancouver BC V6T IZ4 Canada

Rate-splitting multiple access (RSMA) and reinforcement-learning (RL) based user clustering are leveraged to manage interference of a downlink fog radio access network. In particular, clustering of the user devices (UDs) is considered. The UDs of a given cluster can receive data from a certain fog access point (F-AP) over the same radio resource blocks (RRBs) simultaneously. To this end, RSMA enabled data transmission is exploited at each UD cluster. To manage intra-cluster and inter-cluster interference of the network, we focus on the resource allocation to maximize the sum-rate and to minimize total transmission power of the F-APs in each transmission slot. Towards this goal, we optimize jointly the F-APs' transmit power allocation for RSMA enabled data transmission, clustering of UDs, and assignment of RRBs among the F-APs. The proposed optimization problem is NP-hard, and as a result, a global optimal solution is computationally intractable even with the centralized implementation. To obtain an efficient solution without having the global network information, the proposed optimization problem is decomposed into two sub-problems, namely, UD clustering and resource allocation among the F-APs. Specifically, the UD clusters are obtained by applying a multi-agent RL technique, and the resource allocation among the F-APs is obtained by applying the fractional programming, Lagrangian duality, and alternating optimization techniques. A distributed user clustering-power allocation-RRB assignment (UC-PA-RA) algorithm is proposed, and its convergence to the near-optimal solution is proved. Through extensive simulations, the superiority of the proposed UC-PA-RA algorithm over the contemporary multiple access schemes, UD clustering technique, and RL method is verified.

关键词： Resource management Interference Optimization Downlink signal processing algorithms Silicon carbide Computer architecture Interference management rate-splitting multiple access reinforcement learning user clustering

来源：评论

学校读者我要写书评

暂无评论

Visual Domain Adaptation

引用

IEEE signal processing MAGAZINE 2015年第3期32卷 53-69页

作者： Patel, Vishal M. Gopalan, Raghuraman Li, Ruonan Chellappa, Rama Harvard Univ Cambridge MA 02138 USA Univ Maryland UMD Engn College Pk MD USA Univ Maryland UMD Elect & Commun Engn Dept College Pk MD USA Int Assoc Pattern Recognit Adelaide SA Australia Opt Soc Amer Washington DC USA Amer Assoc Advancement Sci Cambridge MA USA Assoc Comp Machinery New York NY USA

In pattern recognition and computer vision, one is often faced with scenarios where the training data used to learn a model have different distribution from the data on which the model is applied. Regardless of the cause, any distributional change that occurs after learning a classifier can degrade its performance at test time. Domain adaptation tries to mitigate this degradation. In this article, we provide a survey of domain adaptation methods for visual recognition. We discuss the merits and drawbacks of existing domain adaptation approaches and identify promising avenues for research in this rapidly evolving field.

关键词： Semisupervised learning Classification algorithms Pattern recognition Visualization signal processing algorithms Computer vision Training data

来源：评论

学校读者我要写书评

暂无评论

Internet tomography

引用

IEEE signal processing MAGAZINE 2002年第3期19卷 47-65页

作者： Coates, M Hero, AO Nowak, R Yu, B Univ Michigan Dept Elect Engn & Comp Sci Ann Arbor MI 48109 USA Univ Michigan Dept Biomed Engn Ann Arbor MI 48109 USA Univ Michigan Dept Stat Ann Arbor MI 48109 USA Rice Univ Houston TX 77251 USA Univ Calif Berkeley Berkeley CA 94720 USA

Today's Internet is a massive, distributed network which continues to explode in size as e-commerce and related activities grow. The heterogeneous and largely unregulated structure of the Internet renders tasks such as dynamic routing, optimized service provision, service-level verification, and detection of anomalous/malicious behavior increasingly challenging tasks. The problem is compounded by the fact that one cannot rely on the cooperation of individual servers and routers to aid in the collection of network traffic measurements vital for these tasks. In many ways, network monitoring and inference problems bear a strong resemblance to other "inverse problems" in which key aspects of a system are not directly observable. Familiar signal processing problems such as tomographic image reconstruction, system identification, and array processing all have interesting interpretations in the networking context. This article introduces the new field of network tomography, a field which we believe will benefit greatly from the wealth of signal processing theory and algorithms

关键词： Tomography signal processing algorithms Array signal processing IP networks Web and internet services Routing Network servers Web server Telecommunication traffic Monitoring

来源：评论

学校读者我要写书评

暂无评论

Statistical analysis and spectral estimation techniques for one-dimensional chaotic signals

引用

IEEE TRANSACTIONS ON signal processing 1997年第6期45卷 1495-1506页

作者： Isabelle, SH Wornell, GW MIT ELECT RES LABCAMBRIDGEMA 02139

signals arising out of nonlinear dynamics are compelling models for a wide range of both natural and man-made phenomena, In contrast to signals arising out of linear dynamics, extremely rich behavior is obtained even when we restrict our attention to one-dimensional (1-D) chaotic systems with certain smoothness constraints. An important class of such systems are the so-called Markov maps. We develop several properties of signals obtained from Markov maps and present analytical techniques for computing a broad class of their statistics in closed form, These statistics include, for example, correlations of arbitrary order and all moments of such signals, Among several results, we demonstrate that all Markov maps produce signals with rational spectra, and we can therefore view the associated signals as ''chaotic ARMA processes,'' with ''chaotic white noise'' as a special case, Finally, we also demonstrate how Markov maps can be used to approximate to arbitrary accuracy the statistics any of a broad class of non-Markov chaotic maps.

关键词： Statistical analysis Chaos signal processing signal analysis State-space methods Statistics signal processing algorithms Power system modeling Analog-digital conversion signal generators

来源：评论

学校读者我要写书评

暂无评论

AUTOMATIC VIEWING RANGE TRACING METHOD FOR COMMUNICATION EQUIPMENT

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 1991年第3期37卷 424-432页

作者： YAMAMOTO, K TSUKADA, H MIZUKOSHI, H NIPPON TELEGRAPH & TEL PUBL CORP ELECT COMMUN LAB TOKYO 100 JAPAN NIPPON TELEGRAPH & TEL PUBL CORP DIV DATA COMMUN TOKYO 100 JAPAN

It is becoming very important to improve face-to-face telecommunication using video telephone sets or video conference equipment by further developing telecommunication technology. Towards this end, an automatic viewing range tracing (AVRT) feature, through which the field of vision automatically traces the person in front of a camera, has been developed for still picture video telephone sets, video doorphones etc.. A new algorithm and LSI have been developed to achieve this feature, and a video doorphone adopting this new feature has been achieved.

关键词： Communication equipment Videoconference signal processing algorithms signal processing Telephony Humans Infrared image sensors Cameras Charge-coupled image sensors Large scale integration

来源：评论

学校读者我要写书评

暂无评论

FIRST BREAK REFRACTION EVENT PICKING USING FUZZY-LOGIC SYSTEMS

引用

IEEE TRANSACTIONS ON FUZZY SYSTEMS 1994年第4期2卷 255-266页

作者： CHU, CKP MENDEL, JM Signal and Image Processing Institute Department of Electrical Engineering-Systems University of Southern California Los Angeles CA USA

First break picking is a pattern recognition problem in seismic signal processing, one that requires much human effort and is difficult to automate. Our goal is to reduce the manual effort in the picking process and accurately perform the picking. Recently, feedforward neural network first break pickers have been developed using back propagation training algorithms applied either to an encoded version of the raw data [8] or to derived seismic attributes which are extracted from the raw data [15]. In this paper we summarize a study in which we applied a recently developed back-propagation fuzzy logic system (BPFLS) to first break picking. We use derived seismic attributes as features, and take lateral variations (defined in main body) into account by using the distance to a piecewise linear guiding function as a new feature. Experimental results indicate that the BPFLS achieves about the same picking accuracy as a feedforward neural network that is also trained using a back propagation algorithm;however, the BPFLS is trained in a much shorter time, because there is a systematic way in which the initial parameters of the BPFLS can be chosen, versus the random way in which the weights of the neural network are chosen.

关键词： Fuzzy logic Humans Neural networks signal processing algorithms Feedforward neural networks Pattern recognition Reflection Data mining Piecewise linear techniques

来源：评论

学校读者我要写书评

暂无评论

On the Convergence of Nested Decentralized Gradient Methods With Multiple Consensus and Gradient Steps

引用

IEEE TRANSACTIONS ON signal processing 2021年 69卷 4192-4203页

作者： Berahas, Albert S. Bollapragada, Raghu Wei, Ermin Univ Michigan Dept Ind & Operat Engn Ann Arbor MI 48019 USA Univ Texas Austin Operat Res & Ind Engn Program Austin TX 78712 USA Northwestern Univ Dept Ind Engn Management Sci Dept Elect & Comp Engn Evanston IL 60208 USA

In this paper, we consider minimizing a sum of local convex objective functions in a distributed setting, where the cost of communication and/or computation can be expensive. We extend and generalize the analysis for a class of nested gradient-based distributed algorithms [NEAR-DGD, (Berahas et al., 2019)] to account for multiple gradient steps at every iteration. We show the effect of performing multiple gradient steps on the rate of convergence and on the size of the neighborhood of convergence, and prove R-Linear convergence to the exact solution with a fixed number of gradient steps and increasing number of consensus steps. We test the performance of the generalized method on quadratic functions and show the effect of multiple consensus and gradient steps in terms of iterations, number of gradient evaluations, number of communications and cost.

关键词： Convergence Optimization Linear programming signal processing algorithms Symmetric matrices Eigenvalues and eigenfunctions Convex functions Distributed optimization communication optimization algorithms network optimization

来源：评论

学校读者我要写书评

暂无评论

DCD-Based Joint Sparse Channel Estimation for OFDM in Virtual Angular Domain

引用

IEEE ACCESS 2021年 9卷 102081-102090页

作者： Liao, Mingduo Zakharov, Yuriy Univ York Dept Elect Engn York YO10 5DD N Yorkshire England

Massive Multiple Input Multiple Output (MIMO) is a promising technique for communications due to the high data transmission rate. To harvest the benefit from the massive MIMO, it is necessary to have accurate channel estimates. Such channels often exhibit sparsity in the virtual angular domain. This paper proposes a dichotomous coordinate descent (DCD) based algorithm for joint sparse channel estimation in the virtual angular domain for the orthogonal-frequency-division-multiplexing massive MIMO. We show that compared to the distributed sparsity adaptive matching pursuit algorithm previously proposed for this purpose, the DCD-based algorithm has significantly lower complexity and better channel estimation performance.

关键词： Channel estimation OFDM signal processing algorithms Matching pursuit algorithms Massive MIMO Downlink Transmitting antennas Channel estimation common sparsity compressive sensing dichotomous coordinate descent distributed sparsity adaptive matching pursuit joint sparse recovery massive MIMO virtual angular domain

来源：评论

学校读者我要写书评

暂无评论

Mapping model with Inter-array memory sharing for multidimensional signal processing

Mapping model with Inter-array memory sharing for multidimen...

引用

IEEE International Conference on Computer-Aided Design

作者： Ilie I. Luican Hongwei Zhu Florin Balasa Department of Computer Science University of Illinois Chicago Chicago IL USA ARM Inc. Sunnyvale CA USA Department of Computer Science Southern Utah University Cedar UT USA

The storage requirements in data-intensive signal processing systems (including applications in video and image processing, artificial vision, medical imaging, real-time 3-D rendering, advanced audio and speech coding) have an important impact on both the system performance and the essential design parameters -the overall power consumption and chip area. This is due to the significant amount of data that must be stored during the execution of the algorithmic specification, as well as due to the amount of data transfers to/from large, energy-consuming, off-chip data memories. This paper addresses the problem of efficiently mapping the multidimensional signals from the algorithmic specification of the system into the physical memory. Different from all the previous mapping models that aim to optimize the memory sharing between the elements of a same array, creating separate windows in the physical memory for distinct arrays, this proposed mapping model is the first one to exploit the possibility of memory sharing between different arrays. As a consequence, this signal-to-memory mapping approach yields significant savings in the amount of data storage resulted after mapping.

关键词： signal mapping Multidimensional signal processing Image storage signal processing algorithms Video signal processing Video sharing Image processing Biomedical imaging Real time systems Rendering (computer graphics)

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：