Skip to content

This was our submission for our Dynamic Programming and Reinforcement Learning (CS6314) Final Project and competition where we won first place. We created 2 models: A Deep Q network and a Rule Based AI. We enlisted in our class competition with the Rule Based AI. Details are in the Document, and Python Notebook.

Notifications You must be signed in to change notification settings

ahmediq-git/3D-Tic-Tac-Toe-AI

Repository files navigation

3D Tic Tac Toe AI

By:

  • Ahmed Mozammil Iqbal (ahmediq-git)
  • Javeria Siddique (Javeria-Siddique)

Summary

This was our submission for our Dynamic Programming and Reinforcement Learning (CS6314) Final Project and competition.

The codebase is divided into 3 phases. The details are given in the Project Report.

The 3rd Phase was our code for the 3D Tic Tac Toe (4 x 4 x 4) Competition where we had to develop an AI or Reinforcement Learning solution that can play as a bot in the game.

We created 2 models:

  • A Deep Q network
  • Rule Based AI

The competition for the 3rd Phase was a tournament tnat involved all groups play their designed AI bots against the other groups AI bots in a game of 3D Tic Tac Toe.

We enlisted in our class competition with the Rule Based AI and won first position in the competition.

Special Thanks to Mahen Mughal (mahen2-cmd) for making the boiler plate code and designing the Project Requirements:

https://github.com/mahen2-cmd/Final-Project-DPRL

About

This was our submission for our Dynamic Programming and Reinforcement Learning (CS6314) Final Project and competition where we won first place. We created 2 models: A Deep Q network and a Rule Based AI. We enlisted in our class competition with the Rule Based AI. Details are in the Document, and Python Notebook.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published