MEReQ: Max-Ent Residual-Q Inverse RL for Sample-Efficient Alignment from Intervention
Chen, Yuxin, Tang, Chen, Li, Chenran, Tian, Ran, Stone, Peter, Tomizuka, Masayoshi, Zhan, Wei
–arXiv.org Artificial Intelligence
Aligning robot behavior with human preferences is crucial for deploying embodied AI agents in human-centered environments. A promising solution is interactive imitation learning from human intervention, where a human expert observes the policy's execution and provides interventions as feedback. However, existing methods often fail to utilize the prior policy efficiently to facilitate learning, thus hindering sample efficiency. In this work, we introduce MEReQ (Maximum-Entropy Residual-Q Inverse Reinforcement Learning), designed for sample-efficient alignment from human intervention. Instead of inferring the complete human behavior characteristics, MEReQ infers a residual reward function that captures the discrepancy between the human expert's and the prior policy's underlying reward functions. It then employs Residual Q-Learning (RQL) to align the policy with human preferences using this residual reward function. Extensive evaluations on simulated and real-world tasks demonstrate that MEReQ achieves sample-efficient policy alignment from human intervention.
arXiv.org Artificial Intelligence
Jun-23-2024
- Country:
- North America > United States
- Texas > Travis County
- Austin (0.04)
- California > Alameda County
- Berkeley (0.05)
- Texas > Travis County
- Asia > China
- Hong Kong (0.04)
- North America > United States
- Genre:
- Research Report
- Promising Solution (0.48)
- New Finding (0.46)
- Research Report
- Technology: