Last semester, I worked on building an AI for K-puzzle, Sudoku and Pacman as part of a team project. For K-puzzle problem, we used A* search with various heuristics, the code can be found here. Of course, we couldn’t use the same idea for Pacman, because the search space was far too big. Instead, we used a reinforcement learning algorithm (Q-learning) for Pacman, the code can be found here.
In the next semester, I’ll be diving deeper into machine learning. I’m interested in implementing a machine learning algorithm for chess. I first got interested in the field of AI after seeing chess AI at work. I’m planning to implement the chess logic and GUI on a web application from scratch, and then build an AI to play the game. This project will be at the intersection of two fields I’m enthusiastic about - web development and artificial intelligence. It’s going to be really interesting :D
