Classifying fMRI images using sparse coding: a project for CS229

Recent work has been done by Rajat Raina and other researchers at Stanford in applying sparse coding techniques to various classification problems. In this project we follow that tradition by applying sparse coding to the problem of classifying fMRI images. In particular, we try to classify fMRI images based on what the subject was doing when the fMRI image was obtained. The two classification possibilities we consider are whether the subject was looking at a sentence or looking at a picture. Rajat Raina provided direction and advise on sparse coding techniques, as well as a code base for sparse coding and classification. We acknowledge him and thank him for his help.