Foundations of Machine Learning

What you’ll be able to do

1. LO1Frame a real problem as a learning task (task, experience, performance measure) and choose a suitable model family.
2. LO2Prepare data and design train, validation and test protocols that avoid leakage.
3. LO3Derive, implement and fit linear models for regression and classification.
4. LO4Diagnose under- and over-fitting and control it with regularisation and cross-validation.
5. LO5Choose, compute and interpret evaluation metrics for regression and classification.
6. LO6Apply and compare non-linear and instance-based methods (k-NN, naive Bayes, SVM, trees, ensembles).
7. LO7Apply unsupervised methods (clustering, dimensionality reduction) and judge their output critically.
8. LO8Run a reproducible end-to-end ML workflow and document a model responsibly.

Chapters

12 chapters · each section ends with a short check

1. 01What it means for a machine to learn
2. 02Data, features and honest evaluation
3. 03Linear regression from first principles
4. 04Generalisation: bias, variance and regularisation
5. 05Logistic regression and probabilistic classification
6. 06Evaluating models
7. 07Nearest neighbours and naive Bayes
8. 08Margins, kernels and linear discriminants
9. 09Decision trees and ensembles
10. 10Clustering
11. 11Dimensionality reduction
12. 12The ML workflow and a first neural network

Graded labs

1. L1

   Linear regression three ways

   After chapter 3 · about 8 hours · Diabetes dataset · BSD-3-Clause (scikit-learn) · Runs on a CPU

2. L2

   Fighting overfitting

   After chapter 4 · about 8 hours · Noisy 1-D curve · Generated for this course; no third-party data · Runs on a CPU

3. L3

   A screening classifier

   After chapter 6 · about 8 hours · Breast Cancer Wisconsin · BSD-3-Clause (scikit-learn) · Runs on a CPU

4. L4

   Trees, forests and boosting

   After chapter 9 · about 8 hours · Wine recognition dataset · BSD-3-Clause (scikit-learn) · Runs on a CPU

5. L5

   Structure without labels

   After chapter 11 · about 8 hours · Optical recognition of handwritten digits · BSD-3-Clause (scikit-learn) · Runs on a CPU

Before you start

Where it leads

How it’s taught

You read short sections written for the page. Each ends with a check, and in-browser exercises let you practise. Graded labs run on your own cloud lab machine, CPU-friendly.

No videos. No live classes. No AI tutor.

How it works→