Work

Inferring bounds on the performance of a control policy from a sample of trajectories

Year: 2009

Type: conference-paper

Abstract: We propose an approach for inferring bounds on the finite-horizon return of a control policy from an off-policy sample of trajectories collecting state transitions, rewards, and control actions. In this paper, the dynamics, control policy, and reward function are supposed to be deterministic and Lipschitz continuous. Under these assumptions, a polynomial algorithm, in terms of the sample size and length of the optimization horizon, is derived to compute these bounds, and their tightness is characterized in terms of the sample density.

Source: IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning

DOI: 10.1109/adprl.2009.4927534

Authors: Raphael Fonteneau, Susan Murphy, Louis Wehenkel, Damien Ernst

Institutions: University of Liège, University of Michigan

Language: English

Topic: Reinforcement Learning in Robotics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Fonds De La Recherche Scientifique - FNRS

OpenAlex ID: W2158013963