Reinforcement Learning with Function-Valued Action Spaces for Partial Differential Equation Control
- Yangchen Pan
- Amir-Massoud Farahmand
- Martha White, University of Alberta
- Saleh Nabi
- Piyush Grover
- Daniel Nikovski
Recent work has shown that reinforcement learning (RL) is a promising approach to control dynamical systems described by partial differential equations (PDE). This paper shows how to use RL to tackle more general PDE control problems that have continuous high-dimensional action spaces with spatial relationship among action dimensions. In particular, we propose the concept of action descriptors, which encode regularities among spatially-extended action dimensions and enable the agent to control high-dimensional action PDEs. We provide theoretical evidence suggesting that this approach can be more sample efficient compared to a conventional approach that treats each action dimension separately and does not explicitly exploit the spatial regularity of the action space. The action descriptor approach is then used within the deep deterministic policy gradient algorithm. Experiments on two PDE control problems, with up to 256-dimensional continuous actions, show the advantage of the proposed approach over the conventional one.
Citation
Y. Pan, A. Farahmand, M. White, S. Nabi, P. Grover, D. Nikovski. "Reinforcement Learning with Function-Valued Action Spaces for Partial Differential Equation Control". International Conference on Machine Learning (ICML), (ed: Jennifer G. Dy, Andreas Krause), pp 3983-3992, July 2018.| Keywords: | |
| Category: | In Conference |
| Web Links: | PMLR |
BibTeX
@incollection{Pan+al:ICML18,
author = {Yangchen Pan and Amir-Massoud Farahmand and Martha White and Saleh
Nabi and Piyush Grover and Daniel Nikovski},
title = {Reinforcement Learning with Function-Valued Action Spaces for
Partial Differential Equation Control},
Editor = {Jennifer G. Dy, Andreas Krause},
Pages = {3983-3992},
booktitle = {International Conference on Machine Learning (ICML)},
year = 2018,
}Last Updated: February 25, 2020Submitted by Sabina P
