Project work for CSE 575 - 'Statistical Machine Learning'.

Problem statement: Training a classifier to predict whether the subject is perceiving a picture or text. The project will focus on successfully applying machine learning methods to train classifiers for data which is as sparse and noisy as it is dimensional. There are 105 features and 6-12 training examples, apparently appropriate feature selection, feature abstraction and classifier training methods tuned to address these characteristics are needed. We will refer the original study to address the aforementioned problem. The classifier will aim to automatically decode the subject’s cognitive state at a single time instant or interval instead of characterizing it over a series of episodes.