Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.09737 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915613359734784 |
|---|---|
| author | Grooten, Bram MacAlpine, Patrick Subramanian, Kaushik Stone, Peter Wurman, Peter R. |
| author_facet | Grooten, Bram MacAlpine, Patrick Subramanian, Kaushik Stone, Peter Wurman, Peter R. |
| contents | Generalization to unseen environments is a significant challenge in the field of robotics and control. In this work, we focus on contextual reinforcement learning, where agents act within environments with varying contexts, such as self-driving cars or quadrupedal robots that need to operate in different terrains or weather conditions than they were trained for. We tackle the critical task of generalizing to out-of-distribution (OOD) settings, without access to explicit context information at test time. Recent work has addressed this problem by training a context encoder and a history adaptation module in separate stages. While promising, this two-phase approach is cumbersome to implement and train. We simplify the methodology and introduce SPARC: single-phase adaptation for robust control. We test SPARC on varying contexts within the high-fidelity racing simulator Gran Turismo 7 and wind-perturbed MuJoCo environments, and find that it achieves reliable and robust OOD generalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_09737 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Out-of-Distribution Generalization with a SPARC: Racing 100 Unseen Vehicles with a Single Policy Grooten, Bram MacAlpine, Patrick Subramanian, Kaushik Stone, Peter Wurman, Peter R. Machine Learning Artificial Intelligence Generalization to unseen environments is a significant challenge in the field of robotics and control. In this work, we focus on contextual reinforcement learning, where agents act within environments with varying contexts, such as self-driving cars or quadrupedal robots that need to operate in different terrains or weather conditions than they were trained for. We tackle the critical task of generalizing to out-of-distribution (OOD) settings, without access to explicit context information at test time. Recent work has addressed this problem by training a context encoder and a history adaptation module in separate stages. While promising, this two-phase approach is cumbersome to implement and train. We simplify the methodology and introduce SPARC: single-phase adaptation for robust control. We test SPARC on varying contexts within the high-fidelity racing simulator Gran Turismo 7 and wind-perturbed MuJoCo environments, and find that it achieves reliable and robust OOD generalization. |
| title | Out-of-Distribution Generalization with a SPARC: Racing 100 Unseen Vehicles with a Single Policy |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2511.09737 |