Reinforcement Learning for poker
Currently implements a moderately accurate simulation of a texas hold'em poker game with a number of automated players and a human player.
Automated players include: dumb - always bets qActor - discrete state Q learning reinforcement learner for preflop betting nnActor - neural network based continuous state Q learning reinforcement learner for preflop betting.