Skip to content
Ahmed Haroon
← Notes

Machine Learning

Supervised and unsupervised learning, learning theory, etc.

  1. 1Linear ModelsLinear regression, logistic regression, the softmax, and their probabilistic foundations.
  2. 2Generative ClassificationGaussian discriminant analysis and Naive Bayes.
  3. 3Kernel MethodsFeature maps, the kernel trick, and what makes a kernel valid.
  4. 4Decision Trees and EnsemblesImpurity, recursive partitioning, bagging, boosting, and AdaBoost.
  5. 5Learning Theory and GeneralizationThe bias-variance tradeoff, complexity bounds, and double descent.
  6. 6Unsupervised Learning and RepresentationClustering data and recovering useful representations with K-means, PCA, and ICA.
  7. 7Expectation Maximization and Mixture ModelsThe ELBO, EM convergence, Gaussian mixtures, and the responsibilities that connect them.
  8. 8Continuous Latent-Variable ModelsFactor analysis, variational inference, and the path from linear latent factors to VAEs.
  9. 9Reinforcement LearningThe agent-environment loop and value function approximation.