Predictive Coding for Boosting Deep Reinforcement Learning with Sparse Rewards
This work addresses sparse reward problems in robotics, which is an incremental improvement over existing methods.
The paper tackles the challenge of sparse rewards in deep reinforcement learning by proposing a reward shaping method using predictive coding, which learns predictive representations offline to provide effective reward signals that understand environment dynamics and emphasize useful features, achieving learning performance comparable to hand-crafted rewards in robotic manipulation and navigation tasks.
While recent progress in deep reinforcement learning has enabled robots to learn complex behaviors, tasks with long horizons and sparse rewards remain an ongoing challenge. In this work, we propose an effective reward shaping method through predictive coding to tackle sparse reward problems. By learning predictive representations offline and using these representations for reward shaping, we gain access to reward signals that understand the structure and dynamics of the environment. In particular, our method achieves better learning by providing reward signals that 1) understand environment dynamics 2) emphasize on features most useful for learning 3) resist noise in learned representations through reward accumulation. We demonstrate the usefulness of this approach in different domains ranging from robotic manipulation to navigation, and we show that reward signals produced through predictive coding are as effective for learning as hand-crafted rewards.