PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training
This work addresses the problem of sample- and feedback-efficient interactive RL for practitioners, offering a novel approach that scales human-in-the-loop methods to more complex tasks.
The paper tackles the challenge of efficiently teaching reinforcement learning agents complex objectives using human feedback by introducing an off-policy interactive RL algorithm that learns a reward model from preference queries and relabels past experience. It demonstrates the method's capability to learn high-complexity tasks like locomotion and robotic manipulation, effectively preventing reward exploitation and enabling behaviors hard to specify with standard rewards.
Conveying complex objectives to reinforcement learning (RL) agents can often be difficult, involving meticulous design of reward functions that are sufficiently informative yet easy enough to provide. Human-in-the-loop RL methods allow practitioners to instead interactively teach agents through tailored feedback; however, such approaches have been challenging to scale since human feedback is very expensive. In this work, we aim to make this process more sample- and feedback-efficient. We present an off-policy, interactive RL algorithm that capitalizes on the strengths of both feedback and off-policy learning. Specifically, we learn a reward model by actively querying a teacher's preferences between two clips of behavior and use it to train an agent. To enable off-policy learning, we relabel all the agent's past experience when its reward model changes. We additionally show that pre-training our agents with unsupervised exploration substantially increases the mileage of its queries. We demonstrate that our approach is capable of learning tasks of higher complexity than previously considered by human-in-the-loop methods, including a variety of locomotion and robotic manipulation skills. We also show that our method is able to utilize real-time human feedback to effectively prevent reward exploitation and learn new behaviors that are difficult to specify with standard reward functions.