Machine learning: Resources

Learn to LaTeX #

Perhaps you too would like to typeset your work using LaTeX. At the very least, it will look correct from twenty feet away. Juntao Lu ‘23 has put together a tutorial that will have you producing great looking mathematics in minutes.

Reading rack #

A collection of articles that complement course material:

  • Computing machinery and intelligence by A. Turing This is the article where Turing introduced the Turing test for intelligence. There are two takeaways: his original proposal is different than what we think of today, and he also believed in ESP and telepathy.

  • A whirlwind linear algebra refresher:

  • A refresher on directional derivatives:

Computational resources #

We will use the Python programming language for the computational aspects of the course.

Google Colab #

The computational aspects of this class will be done primarily in Google’s Colaboratory. By way of an introduction, please read Google’s own blurb about it. It is a wonderful platform to get your feet wet doing machine learning using Python. We will use it as the computational platform for our linear algebra course. You will need a Google account. Computations occur in a notebook which can be simply saved as a Google Doc.

Exercise: Read through Google’s introduction to Colab and then work through this notebook which illustrates how do use Python for some basic computations in linear algebra. You do not need to turn anything in for this exercise.

Bowdoin’s JupyterHub #

For most exercises for our class Google’s Colab will be adequate. But larger projects may require more resources than Google is willing to provide for free. In this case, you may use Bowdoin’s JupyterHub. You will need to either be on campus, or use the Bowdoin VPN service. For instructions, see this link. You may also choose to use the HPC Grid for your computations. See HPC homepage for instructions.

Your own Python installation #

Perhaps you are morally opposed to cloud computing. Or you like to work on the beach far away from an internet signal. Then you should install Python on your own machine. Jeremy Teitelbaum at UConn has put together an excellent set of instructions and an introduction to Python for his machine learning class. Follow his lead.

Direct links to all the Colab notebooks used in the course:

Colab links