论文信息 - A New Algorithm Based On Reinforcement Learning For Regressor Selection In The Least Square Estimation

A New Algorithm Based On Reinforcement Learning For Regressor Selection In The Least Square Estimation

Structure identification means determining the model and it’s details, like the number of regressor in the least square estimation. This is one of the most important problems in system modeling and identification. In this paper a new algorithm based on reinforcement learning is presented for structure identification in least square estimation. The results of the new algorithm are compared with that of the subset selection methods and the steepest descent. Online structure identification is one of the advantages of the proposed method. KeywordsStructure Identification; Least Square Estimation; Reinforcement Learning

Ali Karimpour | Parisa Tavakkoli Heravi

[1] Sheng Chen,et al. Orthogonal least squares methods and their application to non-linear system identification , 1989 .

[2] Kazushi Ikeda,et al. A new criterion using information gain for action selection strategy in reinforcement learning , 2004, IEEE Transactions on Neural Networks.

[3] Shang-Liang Chen,et al. Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[4] O. Nelles. Nonlinear System Identification: From Classical Approaches to Neural Networks and Fuzzy Models , 2000 .

[5] Shie Mannor,et al. Action Elimination and Stopping Conditions for the Multi-Armed Bandit and Reinforcement Learning Problems , 2006, J. Mach. Learn. Res..