DominAI

An AI for the final project of CS221: A dominoes AI

Fall 2016

No files require arguments, so the following are runnable using the following command

python name_of_program.py

• smart_smart.py: PIMC team plays multiple rounds against IMS team
• smart_team.py: An advanced team against greedy. The advanced team is in the smartPlays function, and it is currently PIMC, but can be quickly modified by commenting and uncommenting a few lines of code to become IMS (lines 62-71). The alpha and beta for depth can be changed on line 59, and the number of samples in the calculate_expectation function.
• main.py: Requires two human players, a human to input moves, and a domino set. Plays PIMC team against humans.
• oracle.py: Negamax oracle team against greedy team.

Outputs for smart_smart.py, smart_team.py, and oracle.py can be found in the results_texts folder. We used these for Fig 1 and 2 in our final writeup.

Negamax algorithms used by IMS and PIMC can be found in the algorithms folder. Game and domino classes are in domino.py. The evaluation function, which we modified for one of our runs by uncommenting the latter part of the line, is on line 176 of domino.py.

There are some other files in this repo, that are older iterations of our project or miscellaneous attempts.