Tesseract: Tensorised Actors for Multi-agent Reinforcement Learning

Date:

Talk Slides: In this talk I discuss how tensor decompositions can be used for sample efficient Reinforcement Learning for large factored action spaces.