Machine Learning
for Neuroscience

Interactive companions to the lectures. Every figure is live: drag the points, move the sliders, and watch the numbers follow.

Class 1

Part 1

What is machine learning?

Estimating an unknown function, fitting against predicting, supervised against unsupervised, regression against classification, and looking at your data first.

Open → Part 2

Linear regression

The linear model, least squares, deriving the slope with calculus, R², and what small samples do to all of it.

Open →

Class 2

Interactions, KNN, bias and variance

The price of flexibility

Interactions with a group and between two continuous predictors, nearest-neighbour regression, and the trade that decides how much flexibility is worth paying for.