alphastone
alphastone copied to clipboard
sirmammingtonham
Using self-play, MCTS, and a deep neural network to create a hearthstone ai player
it seems that some cards in CardDefs.xml are not implemented in fireplace,so which vision of CardDefs.xml do you use? the link:https://github.com/HearthSim/hsdata
will overfitting ever be a problem with self-play or would an overfitting neural net always lose to the previous version?
ideas
Embedding vector for all cards with entity_id as index Create additional policy/dqn for mulligans at start of game
Metadata
Owner
Metadata
Using self-play, MCTS, and a deep neural network to create a hearthstone ai player