Learning near-optimal policies with Bellman-residual minimization based fitted policy iteration and a single sample path

Andras Antos 1 Csaba Szepesvari 1 Rémi Munos 2, 3
2 SEQUEL - Sequential Learning
LIFL - Laboratoire d'Informatique Fondamentale de Lille, Inria Lille - Nord Europe, LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
Abstract : We consider batch reinforcement learning problems in continuous space, expected total discounted-reward Markovian Decision Problems. As opposed to previous theoretical work, we consider the case when the training data consists of a single sample path (trajectory) of some behaviour policy. In particular, we do not assume access to a generative model of the environment. The algorithm studied is policy iteration where in successive iterations the Q-functions of the intermediate policies are obtained by means of minimizing a novel Bellman-residual type error. PACstyle polynomial bounds are derived on the number of samples needed to guarantee near-optimal performance where the bound depends on the mixing rate of the trajectory, the smoothness properties of the underlying Markovian Decision Problem, the approximation power and capacity of the function set used.
Document type :
Conference papers
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal.inria.fr/inria-00117130
Contributor : Rémi Munos <>
Submitted on : Thursday, November 30, 2006 - 1:11:01 PM
Last modification on : Thursday, February 21, 2019 - 10:52:49 AM
Long-term archiving on : Tuesday, April 6, 2010 - 11:40:56 PM

File

antos-colt06.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : inria-00117130, version 1

Collections

Citation

Andras Antos, Csaba Szepesvari, Rémi Munos. Learning near-optimal policies with Bellman-residual minimization based fitted policy iteration and a single sample path. Conference On Learning Theory, Jun 2006, Pittsburgh, USA. ⟨inria-00117130⟩

Share

Metrics

Record views

396

Files downloads

1119