Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.03204 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914181678104576 |
|---|---|
| author | Aberdeen, Douglas Baxter, Jonathan |
| author_facet | Aberdeen, Douglas Baxter, Jonathan |
| contents | Policy-gradient methods have received increased attention recently as a mechanism for learning to act in partially observable environments. They have shown promise for problems admitting memoryless policies but have been less successful when memory is required. In this paper we develop several improved algorithms for learning policies with memory in an infinite-horizon setting -- directly when a known model of the environment is available, and via simulation otherwise. We compare these algorithms on some large POMDPs, including noisy robot navigation and multi-agent problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_03204 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Scaling Internal-State Policy-Gradient Methods for POMDPs Aberdeen, Douglas Baxter, Jonathan Machine Learning Policy-gradient methods have received increased attention recently as a mechanism for learning to act in partially observable environments. They have shown promise for problems admitting memoryless policies but have been less successful when memory is required. In this paper we develop several improved algorithms for learning policies with memory in an infinite-horizon setting -- directly when a known model of the environment is available, and via simulation otherwise. We compare these algorithms on some large POMDPs, including noisy robot navigation and multi-agent problems. |
| title | Scaling Internal-State Policy-Gradient Methods for POMDPs |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2512.03204 |