Decision Awareness
in Reinforcement Learning

Workshop at the International Conference on Machine Learning (ICML) 2022

July 22, HALL G T2500 at the Baltimore Convention Center

@DARL_ICML · #DARL_workshop_ICML2022

Summary

This workshop is about decision awareness in RL, which we refer to as the knowledge that each of the components of an RL system should be explicitly trained to help the agent take the optimal action. Classical examples of decision awareness include:

Learning aspects of the dynamics that are important for decision making and not modelling irrelevant parts of the world
Training a critic to directly aid policy optimization instead of focusing on return prediction only
Addressing the sequence of learning problems faced by an agent in its entirety instead of treating them in isolation
Meta-learning algorithmic components that optimize the learning progress itself

For a list of representative papers, please visit the resources page.

Important Dates

Submission site opens	May 1, 2022
Submission deadline	May 27, 2022 AoE
Decisions announced	June 13, 2022
Camera-ready and video submission due	July 1, 2022 AoE
Day of workshop	July 22, 2022