Article (Scientific journals)
Jointly Learning Environments and Control Policies with Projected Stochastic Gradient Ascent
Bolland, Adrien; Boukas, Ioannis; Berger, Mathias et al.
2022In Journal of Artificial Intelligence Research, 73, p. 117-171
Peer reviewed
 

Files


Full Text
DEPS.pdf
Publisher postprint (2.13 MB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
reinforcement learning; joint design and control; deep neural networks; microgrid; drone
Abstract :
[en] We consider the joint design and control of discrete-time stochastic dynamical systems over a finite time horizon. We formulate the problem as a multi-step optimization problem under uncertainty seeking to identify a system design and a control policy that jointly maximize the expected sum of rewards collected over the time horizon considered. The transition function, the reward function and the policy are all parametrized, assumed known and differentiable with respect to their parameters. We then introduce a deep reinforcement learning algorithm combining policy gradient methods with model-based optimization techniques to solve this problem. In essence, our algorithm iteratively approximates the gradient of the expected return via Monte-Carlo sampling and automatic differentiation and takes projected gradient ascent steps in the space of environment and policy parameters. This algorithm is referred to as Direct Environment and Policy Search (DEPS). We assess the performance of our algorithm in three environments concerned with the design and control of a mass-spring-damper system, a small-scale off-grid power system and a drone, respectively. In addition, our algorithm is benchmarked against a state-of-the-art deep reinforcement learning algorithm used to tackle joint design and control problems. We show that DEPS performs at least as well or better in all three environments, consistently yielding solutions with higher returns in fewer iterations. Finally, solutions produced by our algorithm are also compared with solutions produced by an algorithm that does not jointly optimize environment and policy parameters, highlighting the fact that higher returns can be achieved when joint optimization is performed.
Disciplines :
Computer science
Author, co-author :
Bolland, Adrien ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart grids
Boukas, Ioannis  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart-Microgrids
Berger, Mathias ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart grids
Ernst, Damien  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart grids
Language :
English
Title :
Jointly Learning Environments and Control Policies with Projected Stochastic Gradient Ascent
Publication date :
January 2022
Journal title :
Journal of Artificial Intelligence Research
Volume :
73
Pages :
117-171
Peer reviewed :
Peer reviewed
Funders :
F.R.S.-FNRS - Fonds de la Recherche Scientifique [BE]
Available on ORBi :
since 03 June 2020

Statistics


Number of views
554 (53 by ULiège)
Number of downloads
231 (28 by ULiège)

Scopus citations®
 
2
Scopus citations®
without self-citations
1
OpenCitations
 
0

Bibliography


Similar publications



Contact ORBi