Reinforcement Learning with Particle Swarm Optimization Policy (PSO-P) in Continuous State and Action Spaces

Daniel Hein, Alexander Hentschel, Thomas A. Runkler and Steffen Udluft
Additional contact information
Daniel Hein: Technische, Universität München, Munich, Germany
Alexander Hentschel: Siemens AG, Munich, Germany
Thomas A. Runkler: Siemens AG, Munich, Germany
Steffen Udluft: Siemens AG, Munich, Germany

International Journal of Swarm Intelligence Research (IJSIR), 2016, vol. 7, issue 3, 23-42

Abstract: This article introduces a model-based reinforcement learning (RL) approach for continuous state and action spaces. While most RL methods try to find closed-form policies, the approach taken here employs numerical on-line optimization of control action sequences. First, a general method for reformulating RL problems as optimization tasks is provided. Subsequently, Particle Swarm Optimization (PSO) is applied to search for optimal solutions. This Particle Swarm Optimization Policy (PSO-P) is effective for high dimensional state spaces and does not require a priori assumptions about adequate policy representations. Furthermore, by translating RL problems into optimization tasks, the rich collection of real-world inspired RL benchmarks is made available for benchmarking numerical optimization techniques. The effectiveness of PSO-P is demonstrated on the two standard benchmarks: mountain car and cart pole.

Date: 2016
References: Add references at CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://services.igi-global.com/resolvedoi/resolve. ... 018/IJSIR.2016070102 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:igg:jsir00:v:7:y:2016:i:3:p:23-42

Access Statistics for this article

International Journal of Swarm Intelligence Research (IJSIR) is currently edited by Yuhui Shi

More articles in International Journal of Swarm Intelligence Research (IJSIR) from IGI Global
Bibliographic data for series maintained by Journal Editor ().