Excellent. I did want to eventually write something showing how to modify this algorithm to handle games with randomness, and ISMCTS at a glance looks like it does just that. Thanks!
Actually, you don't need information sets to handle games with chance nodes. Standard MCTS works just fine for those. (Don't use a bandit algorithm at the chance nodes, just pick randomly.) Information sets are only needed if you have uncertainty, either because players make simultaneous moves or a player doesn't know (or forgets) the complete state of the game.
