MEReQ: Max-Ent Residual-Q Inverse RL for Sample-Efficient Alignment from Intervention
This work addresses sample efficiency in aligning embodied AI agents with human preferences, which is crucial for deployment in human-centered environments, but it appears incremental as it builds on existing methods like inverse reinforcement learning.
The paper tackles the problem of aligning robot behavior with human preferences through interactive imitation learning from human intervention, introducing MEReQ to infer a residual reward function for more efficient learning, and demonstrates sample-efficient policy alignment in simulated and real-world tasks.
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.