The goal of this assignment is to introduce machine learning literature and practice reading papers. We will also start LaTeX in this assignment and practice writing. On Thursday Oct 13 (updated since Mountain Day) we will discuss these two papers in class, and each student will present a figure (random assignments below).

Paper 1: "Data clustering: 50 years beyond K-means"

Paper Link. This is the HTML version, if you would like to download a PDF but can't access it, or can't access this version, let me know as soon as possible. This is a review paper, which gives an overview of the clustering problem and a variety of challenges and solutions.

Paper 2: "Dynamic hierarchical algorithms for document clustering"

Paper Link. This is the HTML version, if you would like to download a PDF but can't access it, or can't access this version, let me know as soon as possible. This paper presents two specific algorithms, one based on UPGMA, and one based on K-means.

TODO

1. First, if you don't already have it, download LaTeX from this site: LaTeX for all platforms. There is a web option if you don't want to download it or have trouble getting it working on your machine.

2. Read the two papers above and fill out this form for each of them, in LaTeX. I've put the questions in blue, and your responses can be in black. Here is what the form currently looks like as a PDF.

3. Choose one equation from each paper (there aren't too many) that you found useful and reproduce it in LaTeX, then explain what the variables are and what the equation means.

4. Include your figure in your LaTeX document. First save your figure as a PDF or JPG, then uncomment the figure part in the LaTeX document and add the path to where you saved the figure.

   Resources:

   • Detexify is really helpful! Try it out with the "infinity" symbol (∞) or the "element of" symbol (∈).

   • A lot of good equation examples here.

   • Reference Guide that has a lot of useful information.

   Figure Presentations

   Figure assignments are below. Prepare a 1-2 minute explanation of your figure (I will have them on slides, so you don't need to bring anything). Your explanation doesn't need to go into a lot of detail, but try to explain why the figure was included and what the main "takeaway" is. If your figure is less technical (especially true for some of the figures in the first paper), look into the associated references for more detailed information.

   Paper 1

   • Figure 1: Sharon
   • Figure 2: Lujun
   • Figure 3: Sarah
   • Figure 4: Hera
   • Figure 5: Farida
   • Figure 6: Cleo
   • Figure 7: Amelia
   • Figure 8: Jackie
   • Figure 9: Deepshikha
   • Figure 10: Jenny
   • Figure 11: Ravinder
   • Figure 12: Jessica Tran
   • Figure 13: Youyou
   • Figure 14: Isha

   Paper 2

   • Figure 1: Karen
   • Figure 2: Maria
   • Figure 3: Zoe
   • Figure 4: Li
   • Figure 5: Jessica Tin
   • Figure 6: Grace
   • Figure 7: Leah
   • Figure 8: Yvaine
   • Table 2: Alice