Viewing classifier systems as model free learning in POMDPs

被引：0
|
作者：
Hayashi, A [1 ]
Suematsu, N [1 ]
机构：
[1] Hiroshima City Univ, Fac Informat Sci, Asaminami Ku, Hiroshima 7313194, Japan
来源：
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 11 | 1999年 / 11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：
Classifier systems are now viewed disappointing because of their problems such as the rule strength vs rule set performance problem and the credit assignment problem. In order to solve the problems, we have developed a hybrid classifier system: GLS (Generalization Learning System). In designing GLS, we view CSs as model free learning in POMDPs and take a hybrid approach to finding the best generalization, given the total number of rules. GLS uses the policy improvement procedure by Jaakkola et al. for an locally optimal stochastic policy when a set of rule conditions is given. GLS uses GA to search for the best set of rule conditions.
引用
收藏
页码：989 / 995
页数：7
相关论文
共 50 条
  • [21] Risk neutrality in learning classifier systems
    Smith, Justin T. H.
    EVOLUTIONARY INTELLIGENCE, 2012, 5 (02) : 69 - 86
  • [22] Memory Exploitation in Learning Classifier Systems
    Smith, Robert E.
    EVOLUTIONARY COMPUTATION, 1994, 2 (03) : 199 - 220
  • [23] Anticipation mappings for learning classifier systems
    Bull, Larry
    Lanzi, Pier Luca
    O'hara, Toby
    2007 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1-10, PROCEEDINGS, 2007, : 2133 - +
  • [24] An Approach to Active Learning for Classifier Systems
    奚海峰
    罗予频
    杨士元
    JournalofComputerScienceandTechnology, 1999, (04) : 372 - 378
  • [25] Learning cooperation from classifier systems
    Tran, TH
    Sanza, C
    Duthen, Y
    COMPUTATIONAL INTELLIGENCE AND SECURITY, PT 1, PROCEEDINGS, 2005, 3801 : 329 - 336
  • [26] A representational ecology for learning classifier systems
    Marshall, James A. R.
    Kovacs, Tim
    GECCO 2006: GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, VOL 1 AND 2, 2006, : 1529 - +
  • [27] Approach to active learning for classifier systems
    Xi H.
    Luo Y.
    Yang S.
    Journal of Computer Science and Technology, 1999, 14 (4) : 372 - 378
  • [28] Approach to active learning for classifier systems
    Xi, Haifeng
    Luo, Yupin
    Yang, Shiyuan
    Journal of Computer Science and Technology, 1999, 14 (04): : 372 - 378
  • [29] Risk neutrality in learning classifier systems
    Justin T. H. Smith
    Evolutionary Intelligence, 2012, 5 (2) : 69 - 86
  • [30] The 2003 learning classifier systems bibliography
    Kovacs, T
    LEARNING CLASSIFIER SYSTEMS, 2002, 2661 : 187 - 229
12345