Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.06582 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913933389987840 |
|---|---|
| author | Loxley, Peter N. Sommer, Friedrich T. |
| author_facet | Loxley, Peter N. Sommer, Friedrich T. |
| contents | Environments with controllable dynamics are usually understood in terms of explicit models. However, such models are not always available, but may sometimes be learned by exploring an environment. In this work, we investigate using an information measure called "predicted information gain" to determine the most informative regions of an environment to explore next. Applying methods from reinforcement learning allows good suboptimal exploring policies to be found, and leads to reliable estimates of the underlying controllable dynamics. This approach is demonstrated by comparing with several myopic exploration approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_06582 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning controllable dynamics through informative exploration Loxley, Peter N. Sommer, Friedrich T. Machine Learning Artificial Intelligence Environments with controllable dynamics are usually understood in terms of explicit models. However, such models are not always available, but may sometimes be learned by exploring an environment. In this work, we investigate using an information measure called "predicted information gain" to determine the most informative regions of an environment to explore next. Applying methods from reinforcement learning allows good suboptimal exploring policies to be found, and leads to reliable estimates of the underlying controllable dynamics. This approach is demonstrated by comparing with several myopic exploration approaches. |
| title | Learning controllable dynamics through informative exploration |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2507.06582 |