SEIHAI: A Sample-efficient Hierarchical AI for the MineRL Competition
This addresses the problem of reducing environment interactions for reinforcement learning in sparse-reward tasks, with incremental improvements in hierarchical methods.
The paper tackled the challenge of sample-efficient learning in the complex ObtainDiamond task by introducing SEIHAI, a hierarchical AI that leverages human demonstrations and task decomposition, achieving first place in the NeurIPS-2020 MineRL competition.
The MineRL competition is designed for the development of reinforcement learning and imitation learning algorithms that can efficiently leverage human demonstrations to drastically reduce the number of environment interactions needed to solve the complex \emph{ObtainDiamond} task with sparse rewards. To address the challenge, in this paper, we present \textbf{SEIHAI}, a \textbf{S}ample-\textbf{e}ff\textbf{i}cient \textbf{H}ierarchical \textbf{AI}, that fully takes advantage of the human demonstrations and the task structure. Specifically, we split the task into several sequentially dependent subtasks, and train a suitable agent for each subtask using reinforcement learning and imitation learning. We further design a scheduler to select different agents for different subtasks automatically. SEIHAI takes the first place in the preliminary and final of the NeurIPS-2020 MineRL competition.