Learning paths
Machine learning foundations
Read these guides in order to learn the core models of machine learning and the evaluation habits that keep results honest.
Stage 1: The core models
- Machine learning
A practical map of machine learning: core algorithms, evaluation, tabular data and how ML is applied across manufacturing, finance, telecom and logistics. - Supervised learning
Supervised learning explained: how models learn from labeled examples, key concepts, a worked example, limits, and its role in tabular foundation models. - Linear regression
Linear regression predicts a number as a weighted sum of input features. Learn how it is fitted, how to read coefficients, its assumptions and when to use it. - Logistic regression
Logistic regression predicts the probability of a class, such as churn or fraud, from input features. Learn the sigmoid, log-odds and regularisation. - Decision trees
Decision trees learn if-then splits on features to make predictions. Learn how splits are chosen, how to prevent overfitting and why ensembles use them. - Random forest
A random forest averages many decorrelated decision trees trained on random rows and features. Learn how it works, its key settings and when to use it. - Gradient boosting
Gradient boosting builds an ensemble of small trees, each correcting the errors of the previous ones. Learn how it works and why it dominates tabular data. - XGBoost
XGBoost is a regularized gradient-boosting library for tabular data. Learn how it works, how to tune it, and how it relates to tabular foundation models.
Stage 2: Evaluating and improving
- Cross-validation
Cross-validation splits data into folds to train and test a model several times for a reliable performance estimate. Learn k-fold and time-series methods. - Evaluation metrics
Evaluation metrics turn model predictions into numbers you can compare. Learn classification, regression, ranking and probabilistic metrics and their uses. - Bias-variance trade-off
The bias-variance trade-off explains why models underfit or overfit, how expected error splits into parts, and how learning curves help you diagnose and fix it. - Regularization
Regularisation is any change to training that makes a model generalise better to new data, from L1 and L2 penalties to dropout, early stopping and tree limits. - Feature engineering
Feature engineering turns raw data into inputs a model can learn from. Learn transforms, date and text features, target encoding, aggregations and selection. - Hyperparameter tuning
Hyperparameter tuning finds the settings a model cannot learn by itself. Learn grid, random and Bayesian search, Hyperband, nested cross-validation and budgets.
How to use this path
Read the steps in order; each guide links onward to related topics, so you can go deeper wherever you need to. If you already know a stage, skip to the next one. All learning paths are listed on Learning paths.