Task-Agnostic Dynamics Priors for Deep Reinforcement Learning.

Yilun Du, Karthik Narasimhan

Browse the full ICML paper archive.