Machine Learning Fundamentals -
Study Guide
ML workflow, regression, classification, clustering, evaluation and neural
networks - exam-ready notes
Complete Study Guide | Key Concepts | Examples | Practice Questions
, 1. Types of ML and the Standard Workflow
Machine learning learns patterns from data instead of explicit rules. The three paradigms and the workflow below
cover most exam questions.
Type Data Example
Supervised Labelled (x, y) Spam detection, house prices
Unsupervised Unlabelled (x) Customer segmentation
Reinforcement Rewards from environment Game playing, robotics
- Workflow: collect data -> clean -> split train/test -> choose model -> train -> evaluate -> tune -> deploy.
- Never tune on the test set - use a separate validation set or cross-validation.
- Overfitting = memorising noise (high variance); underfitting = too simple (high bias).
- The bias-variance tradeoff is the central tension of model selection.
TIP: Exam one-liner: 'A model that does perfectly on training data but poorly on new data is overfitting.'
Page 1
Study Guide
ML workflow, regression, classification, clustering, evaluation and neural
networks - exam-ready notes
Complete Study Guide | Key Concepts | Examples | Practice Questions
, 1. Types of ML and the Standard Workflow
Machine learning learns patterns from data instead of explicit rules. The three paradigms and the workflow below
cover most exam questions.
Type Data Example
Supervised Labelled (x, y) Spam detection, house prices
Unsupervised Unlabelled (x) Customer segmentation
Reinforcement Rewards from environment Game playing, robotics
- Workflow: collect data -> clean -> split train/test -> choose model -> train -> evaluate -> tune -> deploy.
- Never tune on the test set - use a separate validation set or cross-validation.
- Overfitting = memorising noise (high variance); underfitting = too simple (high bias).
- The bias-variance tradeoff is the central tension of model selection.
TIP: Exam one-liner: 'A model that does perfectly on training data but poorly on new data is overfitting.'
Page 1
