babylonpolice.com

arXiv:2306.06394v5 Announce Type: replace
Abstract: Hierarchical reinforcement learning (HRL) has the potential to solve complex long horizon tasks using temporal abstraction and increased exploration. However, hierarchical agents are difficult to train due to inherent non-stationarity. We present primitive enabled adaptive relabeling (PEAR), a two-phase approach where we first perform adaptive relabeling on a few expert demonstrations to generate efficient subgoal supervision, and then jointly optimize HRL agents by employing reinforcement learning (RL) and imitation learning (IL). We perform theoretical analysis to $(i)$ bound the sub-optimality of our approach, and $(ii)$ derive a generalized plug-and-play framework for joint optimization using RL and IL. Since PEAR utilizes only a handful of expert demonstrations and considers minimal limiting assumptions on the task structure, it can be easily integrated with typical off-policy RL algorithms to produce a practical HRL approach. We perform extensive experiments on challenging environments and show that PEAR is able to outperform various hierarchical and non-hierarchical baselines on complex tasks that require long term decision making. We also perform ablations to thoroughly analyse the importance of our various design choices. Finally, we perform real world robotic experiments on complex tasks and demonstrate that PEAR consistently outperforms the baselines.