What Are Post-Decision States and What Do They Want From Us?

Elucidating transition functions and state-action pairs in Reinforcement Learning

A game of Tic-Tac-Toe perfectly demonstrates the concept of post-decision states [own work by author]

To start with an anti-climax for the seasoned Reinforcement Learning veteran; post-decision states are hardly novel or earth-shattering. Didn’t click away yet? Good, because there is actually some content coming up. In this…