ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918255414738944 |
|---|---|
| author | Zhou, Zihan Garg, Animesh Mandlekar, Ajay Garrett, Caelan |
| author_facet | Zhou, Zihan Garg, Animesh Mandlekar, Ajay Garrett, Caelan |
| contents | Long-horizon manipulation has been a long-standing challenge in the robotics community. We propose ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and motion planning to form an initial solution, and improves each component through reinforcement-learning-based fine-tuning. ReinforceGen first segments the task into multiple localized skills, which are connected through motion planning. The skills and motion planning targets are trained with imitation learning on a dataset generated from 10 human demonstrations, and then fine-tuned through online adaptation and reinforcement learning. When benchmarked on the Robosuite dataset, ReinforceGen reaches 80% success rate on all tasks with visuomotor controls in the highest reset range setting. Additional ablation studies show that our fine-tuning approaches contributes to an 89% average performance increase. More results and videos available in https://reinforcegen.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_16861 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning Zhou, Zihan Garg, Animesh Mandlekar, Ajay Garrett, Caelan Robotics Artificial Intelligence Machine Learning Long-horizon manipulation has been a long-standing challenge in the robotics community. We propose ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and motion planning to form an initial solution, and improves each component through reinforcement-learning-based fine-tuning. ReinforceGen first segments the task into multiple localized skills, which are connected through motion planning. The skills and motion planning targets are trained with imitation learning on a dataset generated from 10 human demonstrations, and then fine-tuned through online adaptation and reinforcement learning. When benchmarked on the Robosuite dataset, ReinforceGen reaches 80% success rate on all tasks with visuomotor controls in the highest reset range setting. Additional ablation studies show that our fine-tuning approaches contributes to an 89% average performance increase. More results and videos available in https://reinforcegen.github.io/ |
| title | ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning |
| topic | Robotics Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2512.16861 |