Learning Process Rewards via Success Visitation Matching for Efficient RL
For robotic control practitioners, this method provides a simple and effective way to accelerate RL finetuning by converting sparse rewards into dense signals without altering the optimal policy.
The paper tackles the problem of sparse rewards in reinforcement learning by proposing a method to transform sparse outcome rewards into dense process rewards using a discriminator trained on successful and unsuccessful episodes. This approach leads to significantly faster RL finetuning on robotic manipulation tasks in both simulation and the real world.
In many modern applications of reinforcement learning (RL), the natural reward for a task of interest is inherently sparse: a reward of 0 is given everywhere except when the task is completed, when a reward of +1 is given. Training a policy to maximize such a sparse reward requires solving a challenging credit assignment problem, leading to slow or ineffective RL improvement. We propose a simple approach to transform a sparse outcome reward into a dense process reward. Our approach relies on training a discriminator to distinguish between previous successful and unsuccessful episodes, and using this discriminator to incentivize the RL-learned policy to match the state-action visitations of successful episodes, while avoiding those of unsuccessful episodes. By incentivizing the policy to match the visitations over all states, not just those that correspond to task success, this reward provides dense feedback on whether progress is being made towards task completion, and, we show, provably achieves this without changing the optimal policy. Focusing on finetuning of robotic control policies, we demonstrate that our approach leads to significantly faster RL finetuning performance on both simulated and real-world manipulation tasks, as compared to simply maximizing the sparse outcome reward.