浏览全部资源
扫码关注微信
1. 苏州大学 计算机科学与技术学院,江苏 苏州 215006
2. 吉林大学 符号计算与知识工程教育部重点实验室,吉林 长春 130012
[ "施梦宇(1989-),男,江苏淮安人,苏州大学硕士生,主要研究方向为强化学习。" ]
[ "刘全(1969-),男,内蒙古牙克石人,苏州大学教授、博士生导师,主要研究方向为强化学习、智能信息处理和自动推理。" ]
[ "傅启明(1985-),男,江苏淮安人,苏州大学博士生,主要研究方向为强化学习、贝叶斯推理和遗传算法。" ]
网络出版日期:2015-02,
纸质出版日期:2015-02-25
移动端阅览
施梦宇, 刘全, 傅启明. 支持合并的自适应tile coding算法[J]. 通信学报, 2015,36(2):186-192.
Meng-yu SHI, Quan LIU, Qi-ming FU. Mergeable adaptive tile coding method[J]. Journal on communications, 2015, 36(2): 186-192.
施梦宇, 刘全, 傅启明. 支持合并的自适应tile coding算法[J]. 通信学报, 2015,36(2):186-192. DOI: 10.11959/j.issn.1000-436x.2015047.
Meng-yu SHI, Quan LIU, Qi-ming FU. Mergeable adaptive tile coding method[J]. Journal on communications, 2015, 36(2): 186-192. DOI: 10.11959/j.issn.1000-436x.2015047.
针对自适应 tile coding 算法会产生多余划分的问题,提出一种支持合并的自适应 tile coding 算法——MATC。该算法能够消除传统自适应tile coding算法中产生的多余划分,进一步解决连续状态空间离散化的问题。将MATC算法应用于离散动作连续状态的Mountain Car问题上,实验结果表明,该算法在学习过程中能消除传统tile coding算法的误划分所产生的不良影响,更准确地自动调整划分的精度,并更快地收敛到最佳策略。
In order to solve many unnecessary division
mergence supported adaptive tile coding algorithm was presented which would eliminate the unnecessary division.Simulation is conducted on mountain car problem with discrete actions and continuous state space Results show that the proposed method can eliminate the influence of false division in the traditional tile coding method and achieve a more accurate adaptive partition of continuous state space.A higher convergence rate is achieved at the same time.
SUTTON R S , BARTO A G . Reinforcement Learning:An Introduction [M ] . Cambridge:MIT Press , 1998 .
LIN C S , KIM H . Selection of learning parameters for CMAC-based adaptive critic learning [J ] . IEEE Trans Neural Networks , 1999 , 6 ( 3 ): 642 - 647 .
PELLEG D , MOORE A , SHROFF N B . X-means:extending K-Means with efficient estimation of the number of clusters [A ] . Proc of the 17th International Conf on Machine Learning [C ] . Boston:Morgan Kaufmann Press , 2000 . 727 - 734 .
PELLEG D , MOORE A . Accelerating exact k-means algorithms with geometric reasoning [A ] . Proc of the fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining [C ] . 1999 . 277 - 281 .
陈宗海 , 文锋 , 聂建斌 等 . 基于节点生长 k-均值聚类算法的强化学习方法 [J ] . 计算机研究与发展 , 2006 , 43 ( 4 ): 661 - 666 .
CHEN Z H , WEN F , NIE J B , et al . A reinforcement learning method based on node-growing k-means cluster algorithm [J ] . Journal of Computer Research and Development , 2006 , 43 ( 4 ): 661 - 666 .
文锋 , 陈宗海 , 卓睿 等 . 连续状态自适应离散化基于K-均值聚类的强化学习方法 [J ] . 控制与决策 , 2006 , 21 ( 2 ): 143 - 147 .
WEN F , CHEN Z H , ZHUO R , et al . Reinforcement learning method of continuous state adaptively discretized based on K-means clustering [J ] . Control and Decision , 2006 , 21 ( 2 ): 143 - 147 .
顾冬雷 , 陈卫东 , 席裕庚 . 一种基于增强学习的自适应控制方法 [J ] . 控制与决策 , 2002 , 17 ( 4 ): 473 - 479 .
GU D L , CHEN W D , XI Y G . A novel adaptive control algorithm based on reinforcement learning [J ] . Control and Decision , 2002 , 17 ( 4 ): 473 - 479 .
MOORE A W , ATKESON C G . The parti-game algorithm for variable resolution reinforcement learning in multidimensional state spaces [J ] . Machine Learning , 1995 , 21 ( 3 ): 199 - 233 .
UTHER W T B , VELOSO M M . Tree based discretization for continuous state space reinforcement learning [A ] . AAAI’98 [C ] . Madison,Wisconsin,United States , 1998
SHERSTOV A A , STONE P . Function Approximation Via Tile Coding:Automating Parameter Choice Abstraction,Reformulation and Approximation [M ] . Springer Berlin Heidelberg , 2005 : 194 - 205 .
WHITESON S , TAYLOR M E , STONE P . Adaptive tile Coding for Value Function Approximation [M ] . Computer Science Department,University of Texas at Austin , 2007 .
WHITESON S , STONE P . Evolutionary function approximation for reinforcement learning [J ] . The Journal of Machine Learning Research , 2006 , 7 : 877 - 917 .
NOKHBEH-ZAEEM M , KHASHABI D , TALEBI H A , et al . Adaptive tiled neural networks [A ] . 2011 IEEE International Conference on Systems,Man,and Cybernetics (SMC) [C ] . New Orleans,LA,USA , 2011 . 2543 - 2548 .
LIN S , WRIGHT R . Evolutionary tile coding:an automated state abstraction algorithm for reinforcement learning [A ] . AAAI Workshops [C ] . 2010 .
0
浏览量
1107
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构