Mihai Anca
Twin delayed hierarchical actor-critic
Anca, Mihai; Studley, Matthew
Authors
Dr Matthew Studley Matthew2.Studley@uwe.ac.uk
Professor of Ethics & Technology/School Director (Research & Enterprise)
Abstract
Hierarchical Reinforcement Learning (HRL) addresses the common problem in sparse rewards environments of having to manually craft a reward function. We present a modified version of the Hierarchical Actor-Critic (HAC) architecture called Twin Delayed HAC (TDHAC), a method capable of sample-efficient learning on environments requiring object interaction. The vanilla algorithm fails to converge on this type of environment, while our method matches the best results so far reported in the literature. We carefully consider each feature added to the original architecture and demonstrate the abilities of TDHAC on the sparse-reward Pick-and-Place environment. To the best of our knowledge, this is the first HRL algorithm successfully applied on an environment requiring object interaction without external enhancements such as demonstrations.
Citation
Anca, M., & Studley, M. (2021). Twin delayed hierarchical actor-critic. In 2021 7th International Conference on Automation, Robotics and Applications (ICARA) (221-225). https://doi.org/10.1109/icara51699.2021.9376459
Conference Name | 2021 International Conference on Automation, Robotics and Applications, ICARA 2021 |
---|---|
Conference Location | Prague, Czech Republic |
Start Date | Feb 4, 2021 |
End Date | Feb 6, 2021 |
Acceptance Date | Dec 25, 2020 |
Publication Date | Mar 17, 2021 |
Deposit Date | Jun 22, 2021 |
Pages | 221-225 |
Book Title | 2021 7th International Conference on Automation, Robotics and Applications (ICARA) |
ISBN | 9780738142906 |
DOI | https://doi.org/10.1109/icara51699.2021.9376459 |
Public URL | https://uwe-repository.worktribe.com/output/7229966 |
You might also like
Trusted research environments for health data
(2024)
Digital Artefact
Achieving goals using reward shaping and curriculum learning
(2023)
Presentation / Conference
Roots and models
(2023)
Journal Article
Arts and humanities shaping the AI future
(2023)
Working Paper
Downloadable Citations
About UWE Bristol Research Repository
Administrator e-mail: repository@uwe.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2024
Advanced Search