Variational Dynamic for Self-Supervised Exploration in Deep Reinforcement Learning

arXiv:2010.08755v3 Announce Type: replace
Abstract: Efficient exploration remains a challenging problem in reinforcement learning, especially for tasks where extrinsic rewards from environments are sparse or even totally disregarded. Significant advances based on intrinsic motivation show promising results in simple environments but often get stuck in environments with multimodal and stochastic dynamics. In this work, we propose a variational dynamic model based on the conditional variational inference to model the multimodality and stochasticity. We consider the environmental state-action transition as a …

abstract advances arxiv cs.cv cs.lg cs.ro dynamic dynamics environments exploration intrinsic motivation multimodal reinforcement reinforcement learning results show simple stochastic tasks type