![]() ![]() The project set out here was named Cepheus. The project is inspired by the work of the students and teachers at the University of Alberta. On an imperfect information environment, to derive results, and hopefully build an intelligent The main purpose of this project is to test applications of deep reinforcement learning methods Contained in this project report will be a discussion about the researchĬarried out in solving the game of Limit Texas Hold’em and a personal pursuit in building a bot Success of researchers sparked global interest and serves as the context for building a pokerīot in today’s climate. Handful have been successful in applying their methods to the game of poker. The world have explored the depths of problems associated with solving such games and only a Poker is recognised as the quintessential game of imperfect information. ![]() Terminal displaying hands being played by the AI bot Tkinter GUI enables users to play against various agent types By default, they are set in both monte_carlo.py and DQN.py as follows: The hyperparameters can be changed from within the text editor if the user wants to disable rendering etc. Run all the cells in the notebook and observe the output. The learner agent is at seat 0 by default.Īlternatively, to review the DQN in google colab, go to this open Jupyter Notebook, and click 'Open in Playground'. A graphical rendering will appear with the game played to 1000 episodes. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |