How to build and run
Run make in the root directory of the repo to generate the program executable main.out.
Run ./main.out white or ./main.out black to start the bot, which follows the tournament communication protocol.
Inspiration
Monte-Carlo Tree Search for board games has a rich history, with famous bots like AlphaGo and AlphaZero using it to achieve state-of-the-art performance against human players in historic matches.
What it does
The bot uses MCTS to search the game tree for the most optimal moves it can find.
How we built it
In C++.
Challenges we ran into
Time constraints prevented us from implementing the entirety of our bot's original design.
Accomplishments that we're proud of
Winning the tournament 🤞
What we learned
Not to leave things to the last minute.
What's next for Our Hex Bot
Augmenting its capabilities with a neural network trained to predict and evaluate board positions, initially by learning from past games played out by others, to eventually learning on its own from scratch.
Log in or sign up for Devpost to join the conversation.