Skip to Main content Skip to Navigation
Documents associated with scientific events

Model-Free Monte Carlo-like Policy Evaluation

Abstract : We propose an algorithm for estimating the finite-horizon expected return of a closed loop control policy from an a priori given (off-policy) sample of one-step transitions. It averages cumulated rewards along a set of ''broken trajectories'' made of one-step transitions selected from the sample on the basis of the control policy. Under some Lipschitz continuity assumptions on the system dynamics, reward function and control policy, we provide bounds on the bias and variance of the estimator that depend only on the Lipschitz constants, on the number of broken trajectories used in the estimator, and on the sparsity of the sample of one-step transitions.
Document type :
Documents associated with scientific events
Complete list of metadata

Cited literature [1 references]  Display  Hide  Download
Contributor : Conférence Mas2010 Connect in order to contact the contributor
Submitted on : Wednesday, August 18, 2010 - 9:42:53 AM
Last modification on : Tuesday, April 21, 2020 - 11:22:33 AM
Long-term archiving on: : Friday, November 19, 2010 - 2:30:28 AM


Files produced by the author(s)


  • HAL Id : inria-00510325, version 1



Raphael Fonteneau, Susan Murphy, Louis Wehenkel, Damien Ernst. Model-Free Monte Carlo-like Policy Evaluation. Journées MAS et Journée en l'honneur de Jacques Neveu, Aug 2010, Talence, France. ⟨inria-00510325⟩



Record views


Files downloads