Assignment One: Multiple Choices

This assignment involved multiple choices.

1. R or whatever other language you want: your choice.
2. Work alone or in pairs, choose a partner: your choice.

And what should you do?  One of the following:

• Use a k-means clustering on the digits in the MNIST dataset, label the clusters according to their most common member, and measure classification performance on the training and testing sets. (Also show the cluster centres, when not using kernel trick).
• Train an SVM to classify the training set (ten SVMs, one SVM for each digit class, with one of them outputting, e.g., +1 for "is a 3" and -1 for "not a 3"), and measure performance on the training and testing sets.

Your choice!


For both the SVM and the k-means above, you should do it twice: once with no kernel trick, and once with the kernel trick.  When using the kernel trick, what kernel(s) should you try?  Your choice!


Each team should turn in their report either via email to barak+cs401@cs.nuim.ie or on paper: your choice.


Due: before class, Mon 1-Nov-2010.

Max vs Min

Oops!  In the lecture of 5-Oct-2010 on finding the maximum margin hyperplane, I wrote max ||w||² where I should have written min ||w||². (Thanks to Thomas Whelan for spotting it.)

Support Vector Machines

Wikipedia has a reasonably good entry on Support Vector Machines. The original paper proposing the technique is also a good resource, quite readable with good motivation: Corinna Cortes and Vladimir N. Vapnik (1995, Support-Vector Networks, Machine Learning 20).