Kernel-based online NEAT for keepaway soccer

被引：0

作者：

Zhao, Yun ^{[1
]}

Cai, Hua ^{[1
]}

Chen, Qingwei ^{[1
]}

Hu, Weili ^{[1
]}

机构：

[1] Nanjing Univ Sci & Technol, Dept Automat, Nanjing 210094, Jiangsu, Peoples R China

来源：

BIO-INSPIRED COMPUTATIONAL INTELLIGENCE AND APPLICATIONS | 2007年 / 4688卷

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a kernel-based online neuroevolutionary of augmenting topology (KO-NEAT) algorithm, which borrowing the selection mechanisms used in temporal difference (TD) algorithms and combining the kernel function approximator for individual fitness initiation. KO-NEAT can improve evolution's online performance of NEAT and learns more quickly. Empirical results in keepaway soccer problem demonstrate that KO-NEAT can substantially improve the original algorithm.

引用

页码：100 / +

页数：2

共 11 条

[1] Reinforcement learning: A survey [J].

Kaelbling, LP ;

Littman, ML ;

Moore, AW .

JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 1996, 4 :237-285

[2]

Stanley KO, 2004, LECT NOTES COMPUT SC, V3103, P1226

[3] Competitive coevolution through evolutionary complexification [J].

Stanley, KO ;

Miikkulainen, R .

JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2004, 21 :63-100

[4] Evolving neural networks through augmenting topologies [J].

Stanley, KO ;

Miikkulainen, R .

EVOLUTIONARY COMPUTATION, 2002, 10 (02) :99-127

[5]

Stone P., 2000, INTEL ROB AUTON AGEN

[6]

Stone P, 2006, LECT NOTES ARTIF INT, V4020, P93

[7]

Sutton R. S., 1988, LEARNING PREDICT MET, P9

[8]

TAYOR ME, 2006, P GEN EV COMP C GECO, P1321

[9]

[童亮 Tong Liang], 2004, [计算机仿真, Computer Simulation], V21, P1

[10]

Whiteson S, 2006, GECCO 2006: GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, VOL 1 AND 2, P1577

← 1 2 →