Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.14972 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912463720546304 |
|---|---|
| author | Huang, Suning Zhang, Zheyu Liang, Tianhai Xu, Yihan Kou, Zhehao Lu, Chenhao Xu, Guowei Xue, Zhengrong Xu, Huazhe |
| author_facet | Huang, Suning Zhang, Zheyu Liang, Tianhai Xu, Yihan Kou, Zhehao Lu, Chenhao Xu, Guowei Xue, Zhengrong Xu, Huazhe |
| contents | Visual deep reinforcement learning (RL) enables robots to acquire skills from visual input for unstructured tasks. However, current algorithms suffer from low sample efficiency, limiting their practical applicability. In this work, we present MENTOR, a method that improves both the architecture and optimization of RL agents. Specifically, MENTOR replaces the standard multi-layer perceptron (MLP) with a mixture-of-experts (MoE) backbone and introduces a task-oriented perturbation mechanism. MENTOR outperforms state-of-the-art methods across three simulation benchmarks and achieves an average of 83% success rate on three challenging real-world robotic manipulation tasks, significantly surpassing the 32% success rate of the strongest existing model-free visual RL algorithm. These results underscore the importance of sample efficiency in advancing visual RL for real-world robotics. Experimental videos are available at https://suninghuang19.github.io/mentor_page/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_14972 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning Huang, Suning Zhang, Zheyu Liang, Tianhai Xu, Yihan Kou, Zhehao Lu, Chenhao Xu, Guowei Xue, Zhengrong Xu, Huazhe Robotics Machine Learning Visual deep reinforcement learning (RL) enables robots to acquire skills from visual input for unstructured tasks. However, current algorithms suffer from low sample efficiency, limiting their practical applicability. In this work, we present MENTOR, a method that improves both the architecture and optimization of RL agents. Specifically, MENTOR replaces the standard multi-layer perceptron (MLP) with a mixture-of-experts (MoE) backbone and introduces a task-oriented perturbation mechanism. MENTOR outperforms state-of-the-art methods across three simulation benchmarks and achieves an average of 83% success rate on three challenging real-world robotic manipulation tasks, significantly surpassing the 32% success rate of the strongest existing model-free visual RL algorithm. These results underscore the importance of sample efficiency in advancing visual RL for real-world robotics. Experimental videos are available at https://suninghuang19.github.io/mentor_page/. |
| title | MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2410.14972 |