Projects per year
Abstract
In many settings, as for example wind farms, multiple machines are instantiated to perform the same task, which is called a fleet. The recent advances with respect to the Internet of Things allow control devices and/or machines to connect through cloud-based architectures in order to share information about their status and environment. Such an infrastructure allows seamless data sharing between fleet members, which could greatly improve the sample-efficiency of reinforcement learning techniques. However in practice, these machines, while almost identical in design, have small discrepancies due to production errors or degradation, preventing control algorithms to simply aggregate and employ all fleet data. We propose a novel reinforcement learning method that learns to transfer knowledge between similar fleet members and creates member-specific dynamical models for control. Our algorithm uses Gaussian processes to establish cross-member covariances. This is significantly different from standard transfer learning methods, as the focus is not on sharing information over tasks, but rather over system specifications. We demonstrate our approach on two benchmarks and a realistic wind farm setting. Our method significantly outperforms two baseline approaches, namely individual learning and joint learning where all samples are aggregated, in terms of the median and variance of the results.
Original language | English |
---|---|
Title of host publication | Proceedings of the 24th European Conference on Artificial Intelligence (ECAI 2020) |
Editors | Giuseppe De Giacomo, Alejandro Catala, Bistra Dilkina, Michela Milano, Senen Barro, Alberto Bugarin, Jerome Lang |
Publisher | IOS Press |
Pages | 1571-1578 |
Number of pages | 8 |
Volume | 325 |
ISBN (Electronic) | 9781643681009 |
ISBN (Print) | 978-1-64368-100-9 |
DOIs | |
Publication status | Published - 24 Aug 2020 |
Event | European Conference on Artificial Intelligence (ECAI 2020) - Santiago De Compostela, Spain Duration: 29 Aug 2020 → 2 Sep 2020 Conference number: 24 http://ecai2020.eu/ |
Publication series
Name | Frontiers in Artificial Intelligence and Applications |
---|
Conference
Conference | European Conference on Artificial Intelligence (ECAI 2020) |
---|---|
Abbreviated title | ECAI |
Country | Spain |
City | Santiago De Compostela |
Period | 29/08/20 → 2/09/20 |
Internet address |
Fingerprint
Dive into the research topics of 'Fleet Control using Coregionalized Gaussian Process Policy Iteration'. Together they form a unique fingerprint.-
VLAAI1: Subsidie: Onderzoeksprogramma Artificiële Intelligentie (AI) Vlaanderen
1/07/19 → 31/12/23
Project: Applied
-
FWOSB27: Robust Fleet-Wide Reinforcement Learning
Nowe, A., Verstraeten, T. & Helsen, J.
1/01/17 → 31/12/20
Project: Fundamental
-
Fleet Control using Coregionalized Gaussian Process Policy Iteration
Timothy Verstraeten (Speaker)
29 Aug 2020 → 8 Sep 2020Activity: Talk or presentation › Talk or presentation at a conference
-
Fleet-Wide Policy Iteration using Gaussian Processes
Timothy Verstraeten (Speaker)
28 Feb 2020Activity: Talk or presentation › Talk or presentation at a workshop/seminar