CDE: Concept-Driven Exploration for Reinforcement Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917320287322112 |
|---|---|
| author | Mao, Le Liu, Andrew H. Zabounidis, Renos Niu, Yanan Kingston, Zachary Campbell, Joseph |
| author_facet | Mao, Le Liu, Andrew H. Zabounidis, Renos Niu, Yanan Kingston, Zachary Campbell, Joseph |
| contents | Intelligent exploration remains a critical challenge in reinforcement learning (RL), especially in visual control tasks. Unlike low-dimensional state-based RL, visual RL must extract task-relevant structure from raw pixels, making exploration inefficient. We propose Concept-Driven Exploration (CDE), which leverages a pre-trained vision-language model (VLM) to generate object-centric visual concepts from textual task descriptions as weak, potentially noisy supervisory signals. Rather than directly conditioning on these noisy signals, CDE trains a policy to reconstruct the concepts via an auxiliary objective, learning general representations of the concepts and using reconstruction accuracy as an intrinsic reward to guide exploration toward task-relevant objects. Across five challenging simulated visual manipulation tasks, CDE achieves efficient, targeted exploration and remains robust to both synthetic errors and noisy VLM predictions. Finally, we demonstrate real-world transfer by deploying CDE on a Franka arm, attaining an 80\% success rate in a real-world manipulation task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_08851 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CDE: Concept-Driven Exploration for Reinforcement Learning Mao, Le Liu, Andrew H. Zabounidis, Renos Niu, Yanan Kingston, Zachary Campbell, Joseph Robotics Intelligent exploration remains a critical challenge in reinforcement learning (RL), especially in visual control tasks. Unlike low-dimensional state-based RL, visual RL must extract task-relevant structure from raw pixels, making exploration inefficient. We propose Concept-Driven Exploration (CDE), which leverages a pre-trained vision-language model (VLM) to generate object-centric visual concepts from textual task descriptions as weak, potentially noisy supervisory signals. Rather than directly conditioning on these noisy signals, CDE trains a policy to reconstruct the concepts via an auxiliary objective, learning general representations of the concepts and using reconstruction accuracy as an intrinsic reward to guide exploration toward task-relevant objects. Across five challenging simulated visual manipulation tasks, CDE achieves efficient, targeted exploration and remains robust to both synthetic errors and noisy VLM predictions. Finally, we demonstrate real-world transfer by deploying CDE on a Franka arm, attaining an 80\% success rate in a real-world manipulation task. |
| title | CDE: Concept-Driven Exploration for Reinforcement Learning |
| topic | Robotics |
| url | https://arxiv.org/abs/2510.08851 |