Our paper “DEIR: Efficient and Robust Exploration through Discriminative-Model-Based Episodic Intrinsic Rewards” has been accepted at IJCAI 2023 (w/ @swan_104 @yujin_tang @alanyttian). DEIR explores more efficiently, especially in partially observable tasks.

Check out https://arxiv.org/abs/2304.10770 for details, and https://github.com/swan-utokyo/deir for the source code.

Tech for Good

As the first author, Shanchuan Wan explained his research in Tech for Good (CNN).