Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.18202 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917353912008704 |
|---|---|
| author | Morihira, Naoki Nahar, Amal Bharadwaj, Kartik Kato, Yasuhiro Hayashi, Akinobu Harada, Tatsuya |
| author_facet | Morihira, Naoki Nahar, Amal Bharadwaj, Kartik Kato, Yasuhiro Hayashi, Akinobu Harada, Tatsuya |
| contents | A central challenge in image-based Model-Based Reinforcement Learning (MBRL) is to learn representations that distill essential information from irrelevant visual details. While promising, reconstruction-based methods often waste capacity on large task-irrelevant regions. Decoder-free methods instead learn robust representations by leveraging Data Augmentation (DA), but reliance on such external regularizers limits versatility. We propose R2-Dreamer, a decoder-free MBRL framework with a self-supervised objective that serves as an internal regularizer, preventing representation collapse without resorting to DA. The core of our method is a redundancy-reduction objective inspired by Barlow Twins, which can be easily integrated into existing frameworks. On DeepMind Control Suite and Meta-World, R2-Dreamer is competitive with strong baselines such as DreamerV3 and TD-MPC2 while training 1.59x faster than DreamerV3, and yields substantial gains on DMC-Subtle with tiny task-relevant objects. These results suggest that an effective internal regularizer can enable versatile, high-performance decoder-free MBRL. Code is available at https://github.com/NM512/r2dreamer. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_18202 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation Morihira, Naoki Nahar, Amal Bharadwaj, Kartik Kato, Yasuhiro Hayashi, Akinobu Harada, Tatsuya Machine Learning Artificial Intelligence Robotics A central challenge in image-based Model-Based Reinforcement Learning (MBRL) is to learn representations that distill essential information from irrelevant visual details. While promising, reconstruction-based methods often waste capacity on large task-irrelevant regions. Decoder-free methods instead learn robust representations by leveraging Data Augmentation (DA), but reliance on such external regularizers limits versatility. We propose R2-Dreamer, a decoder-free MBRL framework with a self-supervised objective that serves as an internal regularizer, preventing representation collapse without resorting to DA. The core of our method is a redundancy-reduction objective inspired by Barlow Twins, which can be easily integrated into existing frameworks. On DeepMind Control Suite and Meta-World, R2-Dreamer is competitive with strong baselines such as DreamerV3 and TD-MPC2 while training 1.59x faster than DreamerV3, and yields substantial gains on DMC-Subtle with tiny task-relevant objects. These results suggest that an effective internal regularizer can enable versatile, high-performance decoder-free MBRL. Code is available at https://github.com/NM512/r2dreamer. |
| title | R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation |
| topic | Machine Learning Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2603.18202 |