Massively Multiagent Minigames for Training Generalist Agents
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909219356147712 |
|---|---|
| author | Choe, Kyoung Whan Sullivan, Ryan Suárez, Joseph |
| author_facet | Choe, Kyoung Whan Sullivan, Ryan Suárez, Joseph |
| contents | We present Meta MMO, a collection of many-agent minigames for use as a reinforcement learning benchmark. Meta MMO is built on top of Neural MMO, a massively multiagent environment that has been the subject of two previous NeurIPS competitions. Our work expands Neural MMO with several computationally efficient minigames. We explore generalization across Meta MMO by learning to play several minigames with a single set of weights. We release the environment, baselines, and training code under the MIT license. We hope that Meta MMO will spur additional progress on Neural MMO and, more generally, will serve as a useful benchmark for many-agent generalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_05071 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Massively Multiagent Minigames for Training Generalist Agents Choe, Kyoung Whan Sullivan, Ryan Suárez, Joseph Artificial Intelligence Machine Learning Multiagent Systems We present Meta MMO, a collection of many-agent minigames for use as a reinforcement learning benchmark. Meta MMO is built on top of Neural MMO, a massively multiagent environment that has been the subject of two previous NeurIPS competitions. Our work expands Neural MMO with several computationally efficient minigames. We explore generalization across Meta MMO by learning to play several minigames with a single set of weights. We release the environment, baselines, and training code under the MIT license. We hope that Meta MMO will spur additional progress on Neural MMO and, more generally, will serve as a useful benchmark for many-agent generalization. |
| title | Massively Multiagent Minigames for Training Generalist Agents |
| topic | Artificial Intelligence Machine Learning Multiagent Systems |
| url | https://arxiv.org/abs/2406.05071 |