Two timescale analysis of the Alopex algorithm for optimization

被引：12

作者：

Sastry, PS ^{[1
]}

Magesh, M

Unnikrishnan, KP

机构：

[1] Indian Inst Sci, Dept Elect Engn, Bangalore 560012, Karnataka, India

[2] GM Corp, R&D Ctr, Warren, MI 48090 USA

来源：

NEURAL COMPUTATION | 2002年 / 14卷 / 11期

关键词：

D O I：

10.1162/089976602760408044

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Alopex is a correlation-based gradient-free optimization technique useful in many learning problems. However, there are no analytical results on the asymptotic behavior of this algorithm. This article presents a new version of Alopex that can be analyzed using techniques of two timescale stochastic approximation method. It is shown that the algorithm asymptotically behaves like a gradient-descent method, though it does not need (or estimate) any gradient information. It is also shown, through simulations, that the algorithm is quite effective.

引用

页码：2729 / 2750

页数：22

共 50 条

[1] An alopex based evolutionary optimization algorithm
Institute of Automation, East China University of Science and Technology, Shanghai 200237, China
[J]. Moshi Shibie yu Rengong Zhineng, 2009, 3 (452-456):
[2] A TWO-TIMESCALE STOCHASTIC ALGORITHM FRAMEWORK FOR BILEVEL OPTIMIZATION: COMPLEXITY ANALYSIS AND APPLICATION TO ACTOR-CRITIC
Hong, Mingyi
Wai, Hoi-To
Wang, Zhaoran
Yang, Zhuoran
[J]. SIAM JOURNAL ON OPTIMIZATION, 2023, 33 (01) : 147 - 180
[3] A Hybrid Evolutionary Algorithm Based on Alopex and Estimation of Distribution Algorithm and Its Application for Optimization
Li, Shaojun
Li, Fei
Mei, Zhenzhen
[J]. ADVANCES IN SWARM INTELLIGENCE, PT 1, PROCEEDINGS, 2010, 6145 : 549 - 557
[4] Optimization of MLP/BP for character recognition using a modified alopex algorithm
Shintani, Hirohito
Akutagawa, Masatake
Nagashino, Hirofumi
Pandya, Abhijit
Kinouchi, Yohsuke
[J]. INTERNATIONAL JOURNAL OF KNOWLEDGE-BASED AND INTELLIGENT ENGINEERING SYSTEMS, 2007, 11 (06) : 371 - 379
[5] Buffer allocation optimization in ATM switching networks using ALOPEX algorithm
Pandya, AS
Sen, E
Hsu, S
[J]. NEUROCOMPUTING, 1999, 24 (1-3) : 1 - 11
[6] A four-timescale algorithm for constrained stochastic optimization of RED
Patro, Rajesh Kumar
Bhatnagar, Shalabh
[J]. PROCEEDINGS OF THE 45TH IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-14, 2006, : 1930 - 1935
[7] Generalization and comparison of Alopex learning algorithm and random optimization method for neural networks
Peng, PY
Sirag, D
[J]. IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE, 1998, : 1147 - 1149
[8] Two-timescale analysis of phugoid mode
Giulietti, F
[J]. JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 2003, 26 (05) : 827 - 830
[9] Chemical process dynamic optimization based on hybrid differential evolution algorithm integrated with Alopex
Qin-qin Fan
Zhao-min Lü
Xue-feng Yan
Mei-jin Guo
[J]. Journal of Central South University, 2013, 20 : 950 - 959
[10] A Two-Timescale Duplex Neurodynamic Approach to Biconvex Optimization
Che, Hangjun
Wang, Jun
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (08) : 2503 - 2514

← 1 2 3 4 5 →