In this session we will make artisanal hand-crafted machine learning model weights from small-batch locally sourced data.

Conveniently for participants of this session, these tiny models are also easier to learn machine learning from. You can start to understand what is happening under the hood and discover how to creatively use machine learning techniques all over the place. If you reset your expectations, they’re also pretty entertaining. And they have broad applications: I made the image on the right using conditional probabilities.

I’ll walk through conditional probabilities and how to create a markov model with code examples. We’ll do an interactive demo. And we’ll wrap up with a discussion about how to apply machine learning in your creative endeavors.

• There will be a bit of math, but it’ll all be counting and multiplication.
• It would be good to have a bit of coding experience. We won’t go too deep into ML (talk to me after!) but it might still be a new perspective for people who have tinkered a bit with machine learning models.
• If you bring a computer, you can follow along with the demos. Otherwise, I’ll post the code on GitHub and you can check it out later.