A practical and theoretical guide to supervised, unsupervised, and ensemble learning—from objective to deployment.
Maps a linear score to a calibrated probability through the sigmoid function.
Measures class mixing inside a decision-tree node; zero means perfectly pure.
Finds the widest separating margin while satisfying the training constraints.
A reusable technical reading framework for moving from vocabulary to working understanding. Use it alongside the formulas and papers below.
A practical and theoretical guide to supervised, unsupervised, and ensemble learning—from objective to deployment. Start by naming the representation, objective, and source of evidence in any system you study. This habit separates a useful model explanation from a list of buzzwords.
Use the learning outcomes as checks for understanding: Select models by task, inductive bias, and data regime; Explain bias–variance trade-offs and evaluation leakage; Compare interpretable baselines with high-capacity models. For each claim, ask what assumptions make it true, what data it needs, and how it could fail.
Translate each equation into a small experiment before treating it as memorized knowledge. Logistic regression: Maps a linear score to a calibrated probability through the sigmoid function. Gini impurity: Measures class mixing inside a decision-tree node; zero means perfectly pure. Support-vector margin: Finds the widest separating margin while satisfying the training constraints.
Read primary work with a repeatable lens: problem, method, data, measurement, limits, and what would change your mind. This section starts with “Random Forests”, “A Tutorial on Support Vector Machines”, “XGBoost: A Scalable Tree Boosting System”. Follow citations outward only after you can explain the central claim in your own words.
Implement a minimal example using Logistic regression; record inputs, outputs, and a failure case.
Define a baseline, a success metric, a stress test, and the evidence required before you trust an improvement.
Choose one paper from the library, reproduce a small claim, then write down the questions its evaluation leaves open.
Primary sources and peer-reviewed work to deepen this page.
Leo Breiman
Defines an ensemble of randomized decision trees with strong accuracy and robust generalization.
C. J. C. Burges
A clear mathematical treatment of maximum-margin classification and kernel methods.
T. Chen · C. Guestrin
Presents a scalable, regularized gradient-boosting system that became a standard tabular baseline.