Too Long; Didn't Read
We’ll be putting out a 3 part blog series giving an introduction to Counterfactual Regret Minimisation (CFR), which is a reinforcement learning algorithm that has recently beaten a number of professional poker players. We’ll be starting with an introduction to the simpler version the algorithm, Regret Matching, (with code) then in the later parts of the series, share some of the findings from our own research and finally share an example of the CFR algorithm that plays a version of poker.