Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.21418 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918169627590656 |
|---|---|
| author | Bierling, Lukas Pasero, Davide Bertrand, Jan-Henrik Van Gerwen, Kiki |
| author_facet | Bierling, Lukas Pasero, Davide Bertrand, Jan-Henrik Van Gerwen, Kiki |
| contents | We introduce DreamerV3-XP, an extension of DreamerV3 that improves exploration and learning efficiency. This includes (i) a prioritized replay buffer, scoring trajectories by return, reconstruction loss, and value error and (ii) an intrinsic reward based on disagreement over predicted environment rewards from an ensemble of world models. DreamerV3-XP is evaluated on a subset of Atari100k and DeepMind Control Visual Benchmark tasks, confirming the original DreamerV3 results and showing that our extensions lead to faster learning and lower dynamics model loss, particularly in sparse-reward settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21418 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DreamerV3-XP: Optimizing exploration through uncertainty estimation Bierling, Lukas Pasero, Davide Bertrand, Jan-Henrik Van Gerwen, Kiki Machine Learning Artificial Intelligence We introduce DreamerV3-XP, an extension of DreamerV3 that improves exploration and learning efficiency. This includes (i) a prioritized replay buffer, scoring trajectories by return, reconstruction loss, and value error and (ii) an intrinsic reward based on disagreement over predicted environment rewards from an ensemble of world models. DreamerV3-XP is evaluated on a subset of Atari100k and DeepMind Control Visual Benchmark tasks, confirming the original DreamerV3 results and showing that our extensions lead to faster learning and lower dynamics model loss, particularly in sparse-reward settings. |
| title | DreamerV3-XP: Optimizing exploration through uncertainty estimation |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2510.21418 |