Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2307.01403 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929231349415936 |
|---|---|
| author | Lo, Yat Long Sengupta, Biswa Foerster, Jakob Noukhovitch, Michael |
| author_facet | Lo, Yat Long Sengupta, Biswa Foerster, Jakob Noukhovitch, Michael |
| contents | Communication is a powerful tool for coordination in multi-agent RL. But inducing an effective, common language is a difficult challenge, particularly in the decentralized setting. In this work, we introduce an alternative perspective where communicative messages sent between agents are considered as different incomplete views of the environment state. By examining the relationship between messages sent and received, we propose to learn to communicate using contrastive learning to maximize the mutual information between messages of a given trajectory. In communication-essential environments, our method outperforms previous work in both performance and learning speed. Using qualitative metrics and representation probing, we show that our method induces more symmetric communication and captures global state information from the environment. Overall, we show the power of contrastive learning and the importance of leveraging messages as encodings for effective communication. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_01403 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Learning Multi-Agent Communication with Contrastive Learning Lo, Yat Long Sengupta, Biswa Foerster, Jakob Noukhovitch, Michael Artificial Intelligence Machine Learning Communication is a powerful tool for coordination in multi-agent RL. But inducing an effective, common language is a difficult challenge, particularly in the decentralized setting. In this work, we introduce an alternative perspective where communicative messages sent between agents are considered as different incomplete views of the environment state. By examining the relationship between messages sent and received, we propose to learn to communicate using contrastive learning to maximize the mutual information between messages of a given trajectory. In communication-essential environments, our method outperforms previous work in both performance and learning speed. Using qualitative metrics and representation probing, we show that our method induces more symmetric communication and captures global state information from the environment. Overall, we show the power of contrastive learning and the importance of leveraging messages as encodings for effective communication. |
| title | Learning Multi-Agent Communication with Contrastive Learning |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2307.01403 |