Exploiting best-match equations for efficient reinforcement learning

MaRDI QIDQ5396671zbMATH OpenFDO

Authors Harm van Seijen, Shimon Whiteson, Hado van Hasselt, Marco A. Wiering

Publication date 3 February 2014

Full work available at URL http://www.jmlr.org/papers/v12/vanseijen11a.html

zbMATH Keywords

data reusefunction approximationon-line learningreinforcement learningtemporal-difference methods

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Markov and semi-Markov decision processes (90C40)

Recommendations

• scientific article; zbMATH DE number 1453042
• Efficient exploration through active learning for value function approximation in reinforcement learning
• Sparse approximations to value functions in reinforcement learning
• Deep exploration via randomized value functions
• Breaking the sample complexity barrier to regret-optimal model-free reinforcement learning

Cited in

(3)

• Probability matching and reinforcement learning
• The Actor-Critic Learning Is Behind the Matching Law: Matching Versus Optimal Behaviors
• Deep reinforcement learning for approximate policy iteration: convergence analysis and a post-earthquake disaster response case study

This page was built for publication: Exploiting best-match equations for efficient reinforcement learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5396671)