You signed in with another tab or window.Reload to refresh your session.You signed out in another tab or window.Reload to refresh your session.You switched accounts on another tab or window.Reload to refresh your session.Dismiss alert
In this project we try to create a sophisticated computer agent to play the Contact Bridge card game. Our goal is to develop an agent that is tough to play against, with fast reaction time so it is able to play in real time against humans. We approached this as a search problem, and implemented search-tree heuristics based on Minimax and Monte C…
In this project we try to create a sophisticated computer agent to play the Contact Bridge card game. Our goal is to develop an agent that is tough to play against, with fast reaction time so it is able to play in real time against humans. We approached this as a search problem, and implemented search-tree heuristics based on Minimax and Monte Carlo Tree Search. Implemented as a final project for the "Introduction to Aritifical Intelligence" course of the Hebrew University of Jerusalem.
Activate virtual environment -source venv/bin/activate.csh if usingtcsh; if usingbash, usesource venv/bin/activate
pip install -r requirements.txt to install project dependencies.
To run a match, runpython3.7 match.py --agent1 <agent arguments> --agent2 <agent arguments> --num_games <int> --verbose_mode <0/1> where each agent encoding is of the form described in match.py’s documentation.
When done, rundeactivate to deactivate the virtual environment
We encourage you to try and run match.py with some of the following arguments preferably from a console outside of an IDE (during the game press the Enter key to perform the next action):
--agent1 MCTS-pure-LowestFirst-250 --agent2 Random
If you wish to run games automatically without seeing each game state and without pressing Enter after each move, add the argument “--verbose_mode 0”.For more arguments, refer tomatch.py.
Authors
Ram Friedman
Yonny Hammer
Maryna Romanchuk
Oriyan Hermoni
About
In this project we try to create a sophisticated computer agent to play the Contact Bridge card game. Our goal is to develop an agent that is tough to play against, with fast reaction time so it is able to play in real time against humans. We approached this as a search problem, and implemented search-tree heuristics based on Minimax and Monte C…