Dream to Adapt: Meta Reinforcement Learning byLatent Context Imagination and MDP Imagination阅读要点

Abstract—Meta reinforcement learning (Meta RL) has been
amply explored to quickly learn an unseen task by transferring
previously learned knowledge from similar tasks. However, most
state-of-the-art Meta RL algorithms require the meta-training
tasks to have a dense coverage of the task distribution and a
great amount of data for each of them. In this paper, we propose
MetaDreamer, a context-based Meta RL algorithm that requires
less real training tasks and data by doing meta-imagination and
MDP-imagination (Markov-Decision-Process). We perform meta-
imagination by interpolating on the learned latent context space
with disentangled properties, as well as MDP-imagination throu