期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

一种改进的强化学习方法在RoboCup中应用研究 被引量：2

Improved Reinforcement Learning Algorithm and Its Application in RoboCup
在线阅读 下载PDF
导出
摘要 基于CMAC(cerebella model articulation controller)提出一种动态强化学习方法(dynamic cerebellamodel articulation controller-advantage learning,DCMAC-AL)。该方法利用advantage(λ)learning计算状态-动作函数,强化不同动作的值函数差异,以避免动作抖动;然后在CMAC函数拟合基础上,利用Bellman误差动态添加特征值,提高CMAC函数拟合的自适应性。同时,在RoboCup仿真平台上对多智能体防守任务(takeaway)进行建模,利用新提出的算法进行学习实验。实验结果表明,DCMAC-AL比采用CMAC的advantage(λ)learning方法有更好的学习效果。 An improved algorithm based on CMAC （cerebella model articulation controller） and named DCMAC-AL is proposed. It uses advantage （λ） learning to calculate the state-action function,emphasizes the differences among action values and shuns action oscillation. It creates novel features based on Bellman error to improve the adaption of CMAC. Besides,it provides a mathematic model for takeaway in RoboCup Soccer Simulation and experiment with DCMAC-AL. The results demonstrate that DCMAC-AL outperforms advantage（λ） learning in regard to learning effort.
作者 程显毅 朱倩
机构地区 南通大学计算机科学与技术学院 江苏大学计算机科学与通信工程学院
出处 《广西师范大学学报（自然科学版）》 CAS 北大核心 2010年第3期99-103,共5页 Journal of Guangxi Normal University:Natural Science Edition
基金 国家自然科学基金资助项目(60702056) 江苏省研究生创新项目(ZX09B2042)
关键词 强化学习 AGENT ROBOCUP CMAC reinforcement learning agent RoboCup CMAC
分类号 TP181 [自动化与计算机技术—控制理论与控制工程]
  • 相关文献

参考文献7

  • 1SUTTON S R, BARTO A G. Reinforcement learning [M]. Cambridge, MA : MIT Press, 1998 : 24-26.
  • 2BAKKER B. Reinforcement learning with long short-term memory[C]//Advances in Neural Information Processing System 14. Cambridge ,MA :MIT Press ,2002 : 987-990.
  • 3PHILIPP W K,SHIE M,DOINA P. Automatic basis function construction for approximate dynamic programming and reinforcement learning[C]//Proceedings of the 23rd International Conference on Machine learning. Cambridge:MIT Press,2006 : 1103-1115.
  • 4高阳,胡景凯,王本年,王冬黎.基于CMAC网络强化学习的电梯群控调度[J].电子学报,2007,35(2):362-365. 被引量：13
  • 5李明爱,焦利芳,郝冬梅,乔俊飞.基于多个并行CMAC神经网络的强化学习方法[J].系统仿真学报,2008,20(24):6683-6685. 被引量：2
  • 6STONE P,SUTTON R S,KUHLMANN G. Reinforcement learning for RoboCup-soceer keepaway[J]. Adaptive Behavior ,2005,13(3) : 165-188.
  • 7ATIL S,TOLEDO C B. A new perspective to the keepaway soccer:the takers (ShortPaper)[C]//ISCEN A,EROGUL U. Proc of 7th Int Conf on Autonomous Agents and Multiagent Systems (AAMAS 2008). Estoril,Portugal: Springer Press, 2008 : 566-569.

二级参考文献18

  • 1Barto A G, Suton S, Anderson C W. Neuron like adaptive elements that can solve difficult learning control problems [J]. IEEE Trans on Systems Man and Cybernetics (ISBN: 0-262-01097-6), 1983, 13(5): 834-846.
  • 2Anderson C W. Learning to control an inverted pendulum using neural networks [J]. IEEE Control System Magazine (S0272-1708), 1989, 9(4): 31-35.
  • 3J C H Watkins. Learning from Delayed Rewards PHD [D]. England: University of Cambridge, 1989.
  • 4Si J N, Wang Y T. On-line learning control by association and reinforcement [J]. IEEE Transactions on Neural Net-works ($1045- 9227), 2001, 12(2): 264-276.
  • 5Lin C T, Lee C S G. Reinforcement structure / parameter learning for neural network-based fuzzy logic control systems [J]. IEEE Transaction on Fuzzy systems (S1063-6706), 1994, 2(1): 46-63.
  • 6Y Sakai, K Kurosawa. Develop of elevator supervisory group control system with artificial intelligence[ J] .Hitachi Review, 1984,33:25 - 30.
  • 7M L Siikonen. Elevator traffic simulafion[J]. Simulation, 1993, 61 : 257 - 267.
  • 8H Ujihara,S Tsuji. The revolutionary AI-2100 elevator-group control system and the new intelligent option series [ J ]. Mitsubishi Electric Advance, 1988,45: 5 - 8.
  • 9H Ujihara, M Amano. The latest elevator group-control system [J]. Mitsubishi Electric Advance, 1994,67:10 - 12.
  • 10Cdtes R H, Barto A G. Elevator group control using multiple reinforcement learning agents[ J ]. Machine Learning, 1998, 33 (2) :235 - 262.

共引文献13

同被引文献32

  • 1李瑾,刘全,杨旭东,杨凯,翁东良.一种改进的平均奖赏强化学习方法在RoboCup训练中的应用[J].苏州大学学报（自然科学版）,2012,28(2):21-26. 被引量：2
  • 2Chen M,Klaus D,Ehsan F.User Manual:RoboCup Soccer ServerManual for Soccer Server Version 7.07 and Later[EB/OL].http://sourceforge.net/projects/sserver/files.
  • 3Stone P,Kuhlmann G,Taylor M E,et al.Keepaway Soccer:from Machine Learning Testbed to Benchmark[M].RoboCup 2005:Robot Soccer World Cup IX .Berlin:Springer Verlag,2006:93-105.
  • 4Stone P,Sutton R S,Kuhlmann G.Reinforcement Learning for RoboCup Soccer Keepaway [J].Adaptive Behavior,2005,13(3):165-188.
  • 5Sutton R S,Barto A G.Reinforcement Learning:an Introduction [M].Cambridge,MA:The MIT Press,2012.
  • 6Taylor M,Stone P,Liu Y.Transfer Learning via Inter-taskMappings for Temporal Difference Learning [J].Journal of Machine Learning Research,2007,8(1):2125-2167.
  • 7Fernández F,García J,Veloso M.Probabilistic Policy Reuse for Inter-task Transfer Learning [J].Robotics and Autonomous Systems,2010,58(7):866-871.
  • 8Fernández F,Veloso M.Probabilistic Policy Reuse in a Rein-forcement Learning Agent[C]∥Nakashima H,Wellman M.AAMAS’06 Proceedings of the Fifth International Joint Conference on Autonomous Agents and Multi-agent Systems.New York:ACM Press,2006:720-727.
  • 9Rummery G A,Niranjan M.On-Line Q-learning using Connectionist Systems[R].Cambridge,England:Cambridge University Engineering Department,1994.
  • 10Walsh T J,Li L,Littman M.Transferring State ions between MDPs[C]∥Proceedings of the ICML’06 Workshop on Structural Knowledge Transfer for Machine Learning.2006.

引证文献2

二级引证文献2

广西师范大学学报（自然科学版）
2010年 第3期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部