Statistics

Introduction to Machine Learning - 10-701/15-781

Slides in Keynote and PDF.

Content

  • Probabilities

    • Bayes rule

    • Dependence, independence, conditional probabilities

    • Naive Bayes classifier

    • Probability estimation

  • Tail bounds

    • Convergence of estimates

    • Chernoff, Hoeffding

    • Chebyshev

    • Gaussian

    • A/B testing and practical tail bounds

  • Kernel density estimation

    • Parzen windows

    • Nearest neighbors

    • Model Selection

    • Watson-Nadaraya regression and classification

  • Exponential families

    • Gaussian, multinomial and Poisson

    • Conjugate distributions and smoothing

    • Parameter estimation

    • Integrating out

Supplementary material

Video