Formal Comparison of Simultaneous Perturbation Stochastic Approximation and Random Direction Stochastic Approximation

被引：2

作者：

Peng, Ducheng ^{[1
]}

Chen, Yiwen ^{[2
]}

Spall, James C. ^{[1
,3
]}

机构：

[1] Johns Hopkins Univ, Appl Math & Stat Dept, Whitehead Hall,3400 North Charles St, Baltimore, MD 21218 USA

[2] Johns Hopkins Univ, Appl Math & Stat Dept, Baltimore, MD USA

[3] JHU, Appl Phys Lab & Res, Baltimore, MD USA

来源：

2023 AMERICAN CONTROL CONFERENCE, ACC | 2023年

关键词：

OPTIMIZATION; ALGORITHMS; CONTROLLER; SYSTEMS;

D O I：

10.23919/ACC55779.2023.10156400

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Stochastic approximation (SA) algorithms can be used in system optimization problems when only noisy measurements of a system are available. This paper formally compares the performance of two popular SA algorithms in a multivariate Kiefer-Wolfowitz setting of simultaneous-perturbation SA (SPSA) and the random-directions SA (RDSA). This paper provides sufficient conditions to demonstrate which algorithm has the smaller asymptotic mean squared error (MSE) and numerically presents comparison of SPSA and RDSA in a test function and a model-free control system. The theory and supporting numerics indicate that SPSA has better efficiency (lower MSE) across a broad range of problem settings.

引用

页码：744 / 749

页数：6

共 50 条

[41] Using normalized simultaneous perturbation stochastic approximation for stable convergence in model-free control scheme [J].

Ahmad, Mohd Ashraf ;

Mustapha, Nik Mohd Zaitul Akmal ;

Nasir, Ahmad Nor Kasruddin ;

Tumari, Mohd Zaidi Mohd ;

Ismail, Raja Mohd Taufika Raja ;

Ibrahim, Zuwairie .

PROCEEDINGS OF 4TH IEEE INTERNATIONAL CONFERENCE ON APPLIED SYSTEM INNOVATION 2018 ( IEEE ICASI 2018 ), 2018, :935-938

[42] Simultaneous Perturbation Stochastic Approximation-Based Consensus for Tracking Under Unknown-But-Bounded Disturbances [J].

Granichin, Oleg ;

Erofeeva, Victoria ;

Ivanskiy, Yury ;

Jiang, Yuming .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2021, 66 (08) :3710-3717

[43] A novel data based control method based upon neural network and simultaneous perturbation stochastic approximation [J].

Dong, Na ;

Chen, Zengqiang .

NONLINEAR DYNAMICS, 2012, 67 (02) :957-963

[44] A STOCHASTIC APPROXIMATION ALGORITHM FOR STOCHASTIC SEMIDEFINITE PROGRAMMING [J].

Gaujal, Bruno ;

Mertikopoulos, Panayotis .

PROBABILITY IN THE ENGINEERING AND INFORMATIONAL SCIENCES, 2016, 30 (03) :431-454

[45] Combining the stochastic counterpart and stochastic approximation methods [J].

Dussault, JP ;

Labrecque, D ;

LEcuyer, P ;

Rubinstein, RY .

DISCRETE EVENT DYNAMIC SYSTEMS-THEORY AND APPLICATIONS, 1997, 7 (01) :5-28

[46] Adaptive System Optimization Using Random Directions Stochastic Approximation [J].

Prashanth, L. A. ;

Bhatnagar, Shalabh ;

Fu, Michael ;

Marcus, Steve .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2017, 62 (05) :2223-2238

[47] Maximum Power Point Tracking of Multi-String Photovoltaic Array via Simultaneous Perturbation Stochastic Approximation [J].

Xiao, Yan ;

Li, Yaoyu ;

Seem, John E. ;

Rajashekara, Kaushik .

PROCEEDINGS OF THE ASME 2013 DYNAMIC SYSTEMS AND CONTROL CONFERENCE (DSCC2013), VOL. 1, 2013,

[48] Discrete non-linear adaptive data driven control based upon simultaneous perturbation stochastic approximation [J].

Dong, Na ;

Wu, Aiguo ;

Chen, Zengqiang .

NONLINEAR DYNAMICS, 2013, 72 (04) :883-894

[49] Regularized Iterative Stochastic Approximation Methods for Stochastic Variational Inequality Problems [J].

Koshal, Jayash ;

Nedic, Angelia ;

Shanbhag, Uday V. .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2013, 58 (03) :594-609

[50] The Curse of Memory in Stochastic Approximation [J].

Lauand, Caio Kalil ;

Meyn, Sean .

2023 62ND IEEE CONFERENCE ON DECISION AND CONTROL, CDC, 2023, :7803-7809

← 1 2 3 4 5 →