← All Articles

Bishop PRML - Ch2. Probability Distributions (0)

Posted on

  • density estimiation

    • frequentist

      • MLE (maximum likelihood estimation)

        • likelihood p(Dθ)p(\mathcal{D}{\mid}\theta) 가 최대값이 되는 모수를 구함
    • Bayesian

      • MAP (maximum a posterior)

        • likelihood * prior p(Dθ)p(θ)p(\mathcal{D}{\mid}\theta)p(\theta) 가 최대값이 되는 모수를 구함
  • conjugate priors

    • posterior 분포가 prior와 동일한 함수 형태가 될 수 있도록 만들어줌
    • multinoial - Dirichlet
    • Gaussian - Gaussian
  • nonparametric density estimation

    • histograms
    • nearest-neighbours
    • kernels
Machine LearningMLBookBishop PRML