This Repo contains comprehensive AI course projects that covers various aspects of artificial intelligence, including search algorithms, game development, Markov decision processes, and deep learning.
- Artificial Intelligence Course - Fall 2024
In this phase, several search algorithms were used to solve the Rubik's Cube problem, including Random Search, Iterative Deepening Depth-First Search (IDS-DFS), A* Search, and Bidirectional Breadth-First Search (BiBFS).
A game of Rollit was developed using Pygame. The game is a multiplayer game where each player takes turns to place pieces on a grid. A minimax algorithm with alpha-beta pruning was used to determine the best move for each player.
In this phase, a Markov Decision Process (MDP) was used to optimize a policy for the Frozen Lake environment. Q-learning was used to learn the optimal policy and promising results were obtained.
In this phase, a neural network was used to play the game of Snake using Q-learning. A deep Q-network was used to approximate the Q-function and promising results were obtained.