Model-Free Reinforcement Learning with Continuous Action in Practice

Thomas Degris 1 Patrick Pilarski 2 Richard Sutton 2
1 Flowers - Flowing Epigenetic Robots and Systems
Inria Bordeaux - Sud-Ouest, U2IS - Unité d'Informatique et d'Ingénierie des Systèmes
2 RLAI
Department of Computing Science [Edmonton]
Abstract : Reinforcement learning methods are often con- sidered as a potential solution to enable a robot to adapt to changes in real time to an unpredictable environment. However, with continuous action, only a few existing algorithms are practical for real-time learning. In such a setting, most effective methods have used a parameterized policy structure, often with a separate parameterized value function. The goal of this paper is to assess such actor-critic methods to form a fully specified practical algorithm. Our specific contributions include 1) developing the extension of existing incremental policy-gradient algorithms to use eligibility traces, 2) an empir- ical comparison of the resulting algorithms using continuous actions, 3) the evaluation of a gradient-scaling technique that can significantly improve performance. Finally, we apply our actor-critic algorithm to learn on a robotic platform with a fast sensorimotor cycle (10ms). Overall, these results constitute an important step towards practical real-time learning control with continuous action.
Document type :
Conference papers
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal.inria.fr/hal-00764281
Contributor : Thomas Degris <>
Submitted on : Wednesday, December 12, 2012 - 4:32:05 PM
Last modification on : Wednesday, July 3, 2019 - 10:48:04 AM
Long-term archiving on : Sunday, December 18, 2016 - 12:21:21 AM

File

DegrisACC2012.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00764281, version 1

Citation

Thomas Degris, Patrick Pilarski, Richard Sutton. Model-Free Reinforcement Learning with Continuous Action in Practice. American Control Conference, Jun 2012, Montreal, Canada. ⟨hal-00764281⟩

Share

Metrics

Record views

311

Files downloads

3884