Lab 8: Machine learning project
Due by 9am, Monday May 4

Starting point code

1. First you need to run setup63 to create a git repository for the lab. If you want to work alone do:

   setup63 labs/08 none

   For this lab you may work with a group of two or three.
   • If you want to work with one partner, then one of you needs to run the following while the other one waits until it finishes.

     setup63 labs/08 partnerUsername

   • If you want to work with two partners, then one of you needs to run the following while the other two wait unitl it finishes.

     setup63 labs/08 partner1Username partner2Username

   Once the script finishes, the other partner(s) should run it on their account.

2. For the next step only one partner should copy over the starting point code.

   cd ~/cs63/labs/08
   cp -r ~meeden/public/cs63/labs/08/* ./
   

   This will copy over the starting point files for this lab.

3. Whether you are working alone or with a partner, you should now add all of the files to your git repo, commit, and push them as shown below.

   git add *
   git commit -m "lab8 start"
   git push
   

4. If you are working with a partner, your partner can now pull the changes in.

   cd ~/cs63/labs/08
   git pull
   

The starting point directory includes just two files: project.tex and README.

In the LaTex file, project.tex, you will describe your project. I have provided a basic structure that you should follow. Feel free to change the section headings, or to add additional sections. Recall that you use pdflatex to convert the LaTex into a pdf file. Here is a template for your paper.

In the README file you will give instructions about how to test the code that you create.

Project ideas

This is an open-ended project where you can choose what machine learning topic you'd like to explore further. Below are some suggestions, but I am open to other options, just talk to me about your ideas first:

• Extend a machine learning approach that we explored in a previous lab:
  • Apply genetic algorithms to a new domain.
  • Apply neural networks to different classification problems.
  • Create your own feature extractors for PacMan and apply approximate Q-learning to figuring out how to deal with ghosts in more complex layouts.
  • Apply approximate Q-learning to a completely different type of problem, perhaps a simple game like Tic-Tac-Toe.
• Implement a machine learning method (such as decision trees or growing neural gas) and test them on a specific problem.
• Use an existing library (such as PyBrain or mlpy) that implements a type of machine learning that we didn't explore and apply it to a specific problem.

Possible data sets you may want to explore:

• UCI Machine Learning Repository
• Kaggle

Submitting your code