Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2309.10908 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909613368016896 |
|---|---|
| author | Wolfe, Alicia P. Diamond, Oliver Goeler-Slough, Brigitte Feuerman, Remi Kisielinska, Magdalena Manfredi, Victoria |
| author_facet | Wolfe, Alicia P. Diamond, Oliver Goeler-Slough, Brigitte Feuerman, Remi Kisielinska, Magdalena Manfredi, Victoria |
| contents | This paper examines a novel type of multi-agent problem, in which an agent makes multiple identical copies of itself in order to achieve a single agent task better or more efficiently. This strategy improves performance if the environment is noisy and the task is sometimes unachievable by a single agent copy. We propose a learning algorithm for this multicopy problem which takes advantage of the structure of the value function to efficiently learn how to balance the advantages and costs of adding additional copies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2309_10908 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Multicopy Reinforcement Learning Agents Wolfe, Alicia P. Diamond, Oliver Goeler-Slough, Brigitte Feuerman, Remi Kisielinska, Magdalena Manfredi, Victoria Multiagent Systems Artificial Intelligence This paper examines a novel type of multi-agent problem, in which an agent makes multiple identical copies of itself in order to achieve a single agent task better or more efficiently. This strategy improves performance if the environment is noisy and the task is sometimes unachievable by a single agent copy. We propose a learning algorithm for this multicopy problem which takes advantage of the structure of the value function to efficiently learn how to balance the advantages and costs of adding additional copies. |
| title | Multicopy Reinforcement Learning Agents |
| topic | Multiagent Systems Artificial Intelligence |
| url | https://arxiv.org/abs/2309.10908 |