Machine Learning

🏷 Supervised Classification

The main toolkit for predicting categories — from probabilistic classifiers to geometric decision boundaries.

9 concepts— start at the top and work your way down
  1. 1

    Logistic Regression

    Modelling the probability of a binary outcome using the sigmoid function — fitting by maximum likelihood or gradient descent.

  2. 2

    Naive Bayes

    A probabilistic classifier that applies Bayes' theorem with the (often unrealistic) assumption that features are conditionally independent given the class.

  3. 3

    Linear Discriminant Analysis

    A classification method that finds the linear combination of features maximising between-class separation relative to within-class scatter.

  4. 4

    Quadratic Discriminant Analysis

    Like LDA but allows each class its own covariance matrix — giving quadratic rather than linear decision boundaries.

  5. 5

    K-Nearest Neighbors

    Classifying or predicting by looking at the k closest training points — the simplest non-parametric method, intuitive yet powerful.

  6. 6

    Decision Trees

    Flowchart-like models that recursively partition the feature space by asking yes/no questions — interpretable but prone to overfitting.

  7. 7

    Ensemble Methods

    Combining many weak models into one strong one — bagging reduces variance, boosting reduces bias, and random forests and stacking blend both ideas.

  8. 8

    Support Vector Machines

    Finding the classification boundary with the widest possible margin to the nearest points of each class — and the kernel trick that lets it draw curved boundaries.

  9. 9

    Geometry of Decision Boundaries

    Comparing classifiers by the shape of the regions they carve out — linear hyperplanes, Voronoi cells, axis-aligned boxes, and curved kernel boundaries.