Learning while Sleeping: Integrating Sleep-Inspired Consolidation with Human Feedback Learning.

Sleep plays a vital role in developmental learning. It allows the brain to consolidate daily learning experiences by replaying the memories accumulated throughout the day. In this work, we take inspiration from sleep and propose the Inverse Forward Offline Reinforcement Model (INFORM), a novel scalable framework that first learns a set of behaviours from human evaluative feedback, then consolidates the learning by applying an offline inverse reinforcement learning to the memorised trajectories. Experimental results demonstrate that INFORM is a feedback-efficient method that effectively learns an optimal policy that aligns with the intended behaviour of the human. A comparative analysis shows that the learnt policies are robust to dynamic changes in the environment and the recovered rewards allow the robot to be autonomous in its learning. Project website: https://sites.google.com/view/inform-framework.