INTRO TO AI
CS540 Projects
In the Spring 2022 semester, I took UW-Madison’s Introduction to
Artificial Intelligence class. Over the course of the semester, I
worked on projects with topics in PCA, hierarchal clustering, linear
regression, neural networks, deep learning, and AI in games, where I
implemented minimax and A* algorithims. Here are a few in more detail.
PCA
This project invloved using linear algebra and PCA (Principal
Component Analysis) to project images. It utilized the Yale face
dataset which is a large set of 32x32 face pictures. The program
would convert the image to number array data, find the covariance
matrix, find the eigenvalues, return a new matrix of eigenvectors,
and display an updated image with a smoother appearance.
Neural Networks
Using deep learning and neural networks, this project was able to
predict labels for various images. The model used images of clothing
items such as t-shirts, pants, jackets, shoes, and boots from a
large dataset as training data. Once the model was finished
training, it could evaluate its effectiveness on itself.
Additionally, a user could input an image of a clothing item, and
the name of the clothing item was effectively predicted.
A* On 7-Tile Puzzle
The 7-tile puzzle has the objective of moving from any given state
of the board, to the board above on the right. The constraints are
that only one tile can be moved at a time to an open space, and they
can only moved one space left, right, up, or down. The secondary
objective was to complete the puzzle in as few moves as possible.
This program provided the optimal solution to solve the puzzle with
the minimal number of moves for any state, using the A* algorithm.