The Value Function Polytope in Reinforcement Learning

Jan 2019

The Value Function Polytope in Reinforcement Learning

Jan 2019

We establish geometric and topological properties of the space of value functions in finite state-action Markov decision processes. Our main contribution is the characterization of the nature of its shape: a general polytope (Aigner et al., 2010). To demonstrate this result, we exhibit several properties of the structural relationship between policies and value functions including the line theorem, which shows that the value functions of policies constrained on all but one state describe a line segment. Finally, we use this novel perspective to introduce visualizations to enhance the understanding of the dynamics of reinforcement learning algorithms.

Reference

Linked Profiles

array(1) { ["wp-wpml_current_language"]=> string(2) "en" }

Mila goes virtual

Starting March 16, 2020, Mila shifts its activities to virtual platforms in order to minimize COVID-19 diffusion.

Read more