Machine Learning Foundations

Progress overview

Course sequence

1. Module 1: Frame the Problem Before Choosing a ModelLearners can decide whether a machine-learning model is a suitable response to a practical problem before they choose an algorithm.
2. Module 2: Understand the Data-Generating ProcessLearners can inspect a tabular dataset before modelling and explain what the rows, labels, features, timing, missing values, proxies, and leakage risks mean.
3. Module 3: Regression and Classification FoundationsLearners can train and explain their first supervised regression and classification baselines using scikit-learn.
4. Module 4: Unsupervised Learning and RepresentationLearners can explore structure in unlabeled data, compare scaling choices, inspect clusters and projections, and write a caution note that avoids inventing causes.
5. Module 5: Generalisation, Splits, and LeakageLearners can identify invalid evaluation, repair leakage, and explain a clean train/validation/test protocol.
6. Module 6: Metrics, Thresholds, and Error AnalysisLearners can choose metrics for a decision, compare thresholds, read confusion-matrix metrics, and write a model-quality note.
7. Module 7: Preprocessing and Reproducible PipelinesLearners can build a clean mixed-feature preprocessing and model pipeline that reruns from a fresh state.
8. Module 8: Model Families and Honest ComparisonLearners can compare bounded candidate models against a dummy baseline and choose the simplest adequate model.
9. Module 9: Responsible ML and Behaviour in UseLearners can write a responsible-use memo with subgroup checks, monitoring signals, human review rules, and stop conditions.
10. Module 10: Capstone and Bridge to Modern AILearners can complete one end-to-end supervised learning study and present conclusions without overstating certainty.

Local learning achievements

Privacy and reset

Use only supplied fictional or synthetic data. Download your work before clearing browser data.