Details
| Original language | English |
|---|---|
| Title of host publication | 2020 IEEE 10th International Conference on Intelligent Systems, IS 2020 - Proceedings |
| Editors | Vassil Sgurev, Vladimir Jotsov, Rudolf Kruse, Mincho Hadjiski |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 551-556 |
| Number of pages | 6 |
| ISBN (electronic) | 9781728154565 |
| Publication status | Published - Aug 2020 |
| Externally published | Yes |
| Event | 10th IEEE International Conference on Intelligent Systems, IS 2020 - Sofia, Bulgaria Duration: 28 Aug 2020 → 30 Aug 2020 |
Publication series
| Name | 2020 IEEE 10th International Conference on Intelligent Systems, IS 2020 - Proceedings |
|---|
Abstract
In this work, we study the capabilities and limitations of forward model learning agents and their applications to motion-control tasks. Forward model learning agents learn to approximate the environment dynamics to apply planning algorithms for action-selection. While previous work has shown that forward model learning agents can efficiently learn to play simple video games, we extend their applicability to domains with continuous state and action spaces. Our experiments show that such agents are quickly able to learn an approximate model of their environment, which suffices to solve several simple motion-control tasks. Comparisons with deep reinforcement learning further highlight the sample efficiency of forward model learning agents.
Keywords
- Deep Reinforcement Learning, Forward Model Learning, Motion Control Tasks
ASJC Scopus subject areas
- Computer Science(all)
- Artificial Intelligence
- Computer Science(all)
- Computer Vision and Pattern Recognition
- Mathematics(all)
- Control and Optimization
- Medicine(all)
- Health Informatics
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
2020 IEEE 10th International Conference on Intelligent Systems, IS 2020 - Proceedings. ed. / Vassil Sgurev; Vladimir Jotsov; Rudolf Kruse; Mincho Hadjiski. Institute of Electrical and Electronics Engineers Inc., 2020. p. 551-556 9199978 (2020 IEEE 10th International Conference on Intelligent Systems, IS 2020 - Proceedings).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Forward Model Learning for Motion Control Tasks
AU - Dockhorn, Alexander
AU - Kruse, Rudolf
PY - 2020/8
Y1 - 2020/8
N2 - In this work, we study the capabilities and limitations of forward model learning agents and their applications to motion-control tasks. Forward model learning agents learn to approximate the environment dynamics to apply planning algorithms for action-selection. While previous work has shown that forward model learning agents can efficiently learn to play simple video games, we extend their applicability to domains with continuous state and action spaces. Our experiments show that such agents are quickly able to learn an approximate model of their environment, which suffices to solve several simple motion-control tasks. Comparisons with deep reinforcement learning further highlight the sample efficiency of forward model learning agents.
AB - In this work, we study the capabilities and limitations of forward model learning agents and their applications to motion-control tasks. Forward model learning agents learn to approximate the environment dynamics to apply planning algorithms for action-selection. While previous work has shown that forward model learning agents can efficiently learn to play simple video games, we extend their applicability to domains with continuous state and action spaces. Our experiments show that such agents are quickly able to learn an approximate model of their environment, which suffices to solve several simple motion-control tasks. Comparisons with deep reinforcement learning further highlight the sample efficiency of forward model learning agents.
KW - Deep Reinforcement Learning
KW - Forward Model Learning
KW - Motion Control Tasks
UR - http://www.scopus.com/inward/record.url?scp=85092707661&partnerID=8YFLogxK
U2 - 10.1109/IS48319.2020.9199978
DO - 10.1109/IS48319.2020.9199978
M3 - Conference contribution
AN - SCOPUS:85092707661
T3 - 2020 IEEE 10th International Conference on Intelligent Systems, IS 2020 - Proceedings
SP - 551
EP - 556
BT - 2020 IEEE 10th International Conference on Intelligent Systems, IS 2020 - Proceedings
A2 - Sgurev, Vassil
A2 - Jotsov, Vladimir
A2 - Kruse, Rudolf
A2 - Hadjiski, Mincho
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 10th IEEE International Conference on Intelligent Systems, IS 2020
Y2 - 28 August 2020 through 30 August 2020
ER -