Name
Title | ||
|---|---|---|
A convergent recursive least squares approximate policy iteration algorithm for multi-dimensional Markov decision process with continuous state and action spaces |
Abstract | ||
|---|---|---|
In this paper, we present a recursive least squares approximate policy iteration (RLSAPI) algorithm for infinite-horizon multi-dimensional Markov decision process in continuous state and action spaces. Under certain problem structure assumptions on value functions and policy spaces, the approximate policy iteration algorithm is provably convergent in the mean. That is to say the mean absolute deviation of the approximate policy value function from the optimal value function goes to zero as successive approximation improves. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/ADPRL.2009.4927527 | ADPRL |
Keywords | Field | DocType |
approximation theory,optimal value function,least squares approximations,convergent recursive least square approximate policy iteration algorithm,approximate policy value function,convergence of numerical methods,decision theory,action space,mean absolute deviation,markov processes,iterative methods,multidimensional markov decision process,continuous state space,function approximation,least squares approximation,linear approximation,markov decision process,approximation algorithms,convergence,value function | Least squares,Approximation algorithm,Mathematical optimization,Markov process,Function approximation,Algorithm,Markov decision process,Q-learning,Bellman equation,Recursive least squares filter,Mathematics | Conference |
ISBN | Citations | PageRank |
978-1-4244-2761-1 | 4 | 0.50 |
References | Authors | |
17 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jun Ma | 1 | 93 | 9.42 |
| Warren B. Powell | 2 | 1614 | 151.46 |