Learning from Trajectories via Subgoal Discovery
This addresses the sample inefficiency in reinforcement learning for sparse-reward tasks, offering a novel hybrid approach that could benefit robotics and autonomous systems, though it appears incremental in combining existing techniques.
The paper tackles the problem of learning complex goal-oriented tasks with sparse rewards by decomposing expert trajectories into sub-goals, enabling a switch from imitation learning to reinforcement learning, and achieves success on tasks where prior methods fail.
Learning to solve complex goal-oriented tasks with sparse terminal-only rewards often requires an enormous number of samples. In such cases, using a set of expert trajectories could help to learn faster. However, Imitation Learning (IL) via supervised pre-training with these trajectories may not perform as well and generally requires additional finetuning with expert-in-the-loop. In this paper, we propose an approach which uses the expert trajectories and learns to decompose the complex main task into smaller sub-goals. We learn a function which partitions the state-space into sub-goals, which can then be used to design an extrinsic reward function. We follow a strategy where the agent first learns from the trajectories using IL and then switches to Reinforcement Learning (RL) using the identified sub-goals, to alleviate the errors in the IL step. To deal with states which are under-represented by the trajectory set, we also learn a function to modulate the sub-goal predictions. We show that our method is able to solve complex goal-oriented tasks, which other RL, IL or their combinations in literature are not able to solve.