CSCE 883: Problem Set 1

Hands-on Machine Learning

For this problem set you will implement your own machine learning algorithm and test it with at least three datasets from the UCI Repository. I recommend you try the "Molecular Biology Databases" if you are into biology. You will choose any one of the algorithm that have been taught in class before the Spring Break.

Your implementation you must use some method to avoid overfitting.

What to hand in

1. The well-commented source code.
2. A writeup that includes graphs of the training, and test errors for your three chosen datasets and
3. an analysis of what worked, when, and why, and
4. a timing analysis: how long did it take to learn? to run?

Submission Instructions

As will all the problem sets, you will hand them in using our department's dropbox.