Programmatic Reinforcement Learning for Trustworthy Microgrid Management

Subrat Prasad Panda, Blaise Genest, Arvind Easwaran

Keywords: Trustworthy Reinforcement Learning, Programmatic Policies, Decision Trees, DTSemNet, Microgrid

TL;DR: We propose using DTSemNet, a ∂PRL approach, for microgrid energy management, improving trustworthiness while maintaining performance similar to NNs. This advances the goal of injecting knowledge and discovering causal relationships in energy systems.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 186