Timezone: »

 
Maximum State Entropy Exploration using Predecessor and Successor Representations
Arnav Kumar Jain · Lucas Lehnert · Irina Rish · Glen Berseth

@
in New Frontiers in Learning, Control, and Dynamical Systems »
Event URL: https://openreview.net/forum?id=inE5hW4tQ0 »
Animals have a developed ability to explore that aids them in important tasks such as locating food, exploring for shelter, and finding misplaced items. These exploration skills necessarily track where they have been so that they can plan for finding items with relative efficiency. Contemporary exploration algorithms often learn a less efficient exploration strategy because they either condition only on the current state or simply rely on making random open-loop exploratory moves. In this work, we propose $\eta\psi$-Learning, a method to learn efficient exploratory policies by conditioning on past episodic experience to make the next exploratory move. Specifically, $\eta\psi$-Learning learns an exploration policy that maximizes the entropy of the state visitation distribution of a single trajectory. Furthermore, we demonstrate how variants of the predecessor representation and successor representations can be combined to predict the state visitation entropy. Our experiments demonstrate the efficacy of the proposed algorithm to strategically explore the environment and maximize the state coverage with limited samples.

Author Information

Arnav Kumar Jain (Mila - Quebec AI Institute, University of Montreal)
Lucas Lehnert (Meta FAIR)
Irina Rish (MILA / Université de Montréal h)
Glen Berseth (Mila Universite de Montreal)

More from the Same Authors