π2vec: Policy Representations with Successor Features

06/16/2023
by   Gianluca Scarpellini, et al.
0

This paper describes π2vec, a method for representing behaviors of black box policies as feature vectors. The policy representations capture how the statistics of foundation model features change in response to the policy behavior in a task agnostic way, and can be trained from offline data, allowing them to be used in offline policy selection. This work provides a key piece of a recipe for fusing together three modern lines of research: Offline policy evaluation as a counterpart to offline RL, foundation models as generic and powerful state representations, and efficient policy selection in resource constrained environments.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset