Skip to Main content Skip to Navigation
New interface
Conference papers

Local Policy Search in a Convex Space and Conservative Policy Iteration as Boosted Policy Search

Bruno Scherrer 1 Matthieu Geist 2 
1 MAIA - Autonomous intelligent machine
Inria Nancy - Grand Est, LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
Abstract : Local Policy Search is a popular reinforcement learning approach for handling large state spaces. Formally, it searches locally in a parameterized policy space in order to maximize the associated value function averaged over some pre-defined distribution. The best one can hope in general from such an approach is to get a local optimum of this criterion. The first contribution of this article is the following surprising result: if the policy space is convex, any (approximate) local optimum enjoys a global performance guarantee. Unfortunately, the convexity assumption is strong: it is not satisfied by commonly used parameterizations and designing a parameterization that induces this property seems hard. A natural so-lution to alleviate this issue consists in deriving an algorithm that solves the local policy search problem using a boosting approach (constrained to the convex hull of the policy space). The resulting algorithm turns out to be a slight generalization of conservative policy iteration; thus, our second contribution is to highlight an original connection between local policy search and approximate dynamic pro-gramming.
Complete list of metadata

Cited literature [25 references]  Display  Hide  Download
Contributor : Bruno Scherrer Connect in order to contact the contributor
Submitted on : Monday, December 8, 2014 - 3:30:55 PM
Last modification on : Friday, November 18, 2022 - 9:28:20 AM
Long-term archiving on: : Monday, March 9, 2015 - 10:10:49 AM


Files produced by the author(s)



Bruno Scherrer, Matthieu Geist. Local Policy Search in a Convex Space and Conservative Policy Iteration as Boosted Policy Search. ECML, Sep 2014, Nancy, France. pp.35 - 50, ⟨10.1007/978-3-662-44845-8_3⟩. ⟨hal-01091079⟩



Record views


Files downloads