Fall semester 2013

Since my last post, I’ve completed my the first half of my final year as an undergraduate at Brown University and what a year it has been already! I started the school year off in top gear by competing in a half-Ironman relay and a sprint triathlon.

World's End in Hingham, MA

Before classes really kicked off, I also got a chance to do a quick hike at the dramatically titled World’s End Park in Hingham, Massachusetts. After that though, classes really did pick up and between classwork and applying to graduate school, my spare time became quite limited. My favorite class of the fall semester was Machine Learning, taught by Prof. Erik Sudderth.

Neural network decision boundaries

To start, we learned about the basics of machine learning such as Naive Bayes and maximum likelihood and then moved on to more advanced topics including: linear and logistic regression, regularization and sparsity, neural networks, clustering and expectation-maximization algorithms, Hidden Markov Models, support vector machines, and factor analysis.  Above is a picture of the decision boundaries produced by neural networks with varying numbers of hidden units I produced for one of the homework assignments.

CT Scan of electrode array Electrode locations for thesis

Besides classwork, I also worked on my honors thesis, entitled Neural Dynamics of Natural Vision. I spent many hours in Rhode Island Hospital working on building the experimental apparatus. Pictured above are the electrode locations for the arrays in our subjects.

Despite a stressful semester, I’m more than excited for my final semester as an undergraduate. Let’s hope it’s a good one!

Leave a Reply

Your email address will not be published. Required fields are marked *