We will read the following paper in class together on low dimensional projections and EM. This is to illustrate how to read ML papers for your research. This is short, well written, has a nice result, gives good background and just challenging enough that it is worth it. See the assessment where you will answer questions on this and hopefully tell you how to look at these papers better.

A detailed presentation, with simulations, can be a choice for your final project. If you do this, you will be

  1. Understanding and summarizing the mathematical details
  2. Implementing the algorithm on a dataset I will provide.