Rubin, J., Watson, I.: Computer poker: a review. The best model configuration for predicting the moves of the poker game with three layers with Mean Squared Error (MSE) loss function, Adam (0.001) optimizer, relu activation in the first layer had, relu6 activation in the second layer and selu activation in the third layer has achieved an accuracy of 99%. We tested various combinations of model parameters: number of model layers, activation, loss function, optimization function, and number of epochs. ![]() The model was trained on UCI Poker Hand dataset. We created a sequential model of poker move prediction using TensorFlow.js machine learning library and its sequential model. Therefore, poker is an interesting case used to check the abilities of machine learning models to make optimal decisions. all game-related information is unknown until the end of the game. Poker is attributed to the group of games with incomplete information, i.e. We analyze neural network algorithms that can play poker game.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |