CiteSeerX — Policy Gradient in Continuous Time

Policy Gradient in Continuous Time (2006) [9 citations — 0 self]

Download:
http://www.jmlr.org/papers/volume7/munos06b/munos0
DBLP
CACHED:

by Remi Munos , Michael Littman

Journal of Machine Learning Research

Add To MetaCart

Popular Tags

No tags have been applied to this document.

BibTeX | Add To MetaCart

@ARTICLE{Munos06policygradient,
    author = {Remi Munos and Michael Littman},
    title = {Policy Gradient in Continuous Time},
    journal = {Journal of Machine Learning Research},
    year = {2006},
    volume = {7},
    pages = {771--791}
}

Bookmarks

OpenURL

Abstract:

Policy search is a method for approximately solving an optimal control problem by performing a parametric optimization search in a given class of parameterized policies. In order to process a local optimization technique, such as a gradient method, we wish to evaluate the sensitivity of the performance measure with respect to the policy parameters, the so-called policy gradient. This paper is concerned with the estimation of the policy gradient for continuous-time, deterministic state dynamics, in a reinforcement learning framework, that is, when the decision maker does not have a model of the state dynamics.

Citations

1933	Reinforcement Learning: An introduction – Sutton, Barto - 1998
267	Numerical Solution of Stochastic Differential Equations. 3rd revised printing – Kloeden, Platen - 1999
207	Stochastic Approximation Algorithms and Applications – Kushner, Yin - 1997
188	Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning – Williams - 1992
176	Y.: Policy gradient methods for reinforcement learning with function approximation – Sutton, McAllester, et al. - 2000
162	Planning Algorithms – LAVALLE - 2006
79	The concentration of measure phenomenon – Ledoux - 2001
45	A new look at independence – Talagrand - 1996
27	Introduction to Optimization, Optimization – Polyak - 1987
24	Infinite-horizon gradient-based policy search – Baxter, Bartlett
24	Sensitivity analysis via likelihood ratios – Reiman, Weiss - 1986
6	Sensitivity analysis using Itô Malliavin calculus and martingales. Application to stochastic optimal control – Gobet, Munos - 2002
5	Tsitsiklis. Approximate gradient methods in policy-space optimization of Markov reward processes – Marbach, N - 2003
2	Perturbation methods in optimal control. Wiley/Gauthier-Villars Series in Modern Applied Mathematics – Bensoussan - 1988
1	Optimal control of a double inverted pendulum on a cart – Bogdanov - 2004