In this project, you will design three classifiers: a perceptron classifier, a large-margin (MIRA) classifier, and a slightly modified perceptron classifier for behavioral cloning. You will test the first two classifiers on a set of scanned handwritten digit images, and the last on sets of recorded pacman games from various agents. Even with simple features, your classifiers will be able to do quite well on these tasks when given enough training data.
Python 100.00%
ai-pacman-classification's Introduction
Name: Austin Balogh
CSU ID: 2597975
My implementation of the project involves finding ranges of
data and labels. Then returning a max or min depending on
what is needed or performing math where it is needed.
Thoughout the project the counter class functions are used to
update weights with the proper math needed for them. Much like
previous projects, the pacman enhanced features involve finding
distances to ghosts and food. My MIRA implementation seems to
somewhat long to complete due to the amount of loops used for
the grid. There may be some way to optimize it, but I am not
sure how. I was uncertain how to do question 4, so I did not
submit it. I spent about 25 hours on this project.