Gexin Huang

Gexin Huang

Pattern Recognition, Bayesian Deep Learning, Generative Model, and Multi-Modal learning

Note about Bayesian Machine Learning

less than 1 minute read

Published:

This is a learning note about Bayesian Machine Learning. Their bedrock is Probabilistic Graphic Model (PGMs) and Bayesian Inference.

Road Map

  • Posterior is feasible
    • Maximum a posteriori estimation
    • Expectation Maximization (EM) approximation
  • Posterior is intractable
    • Sampling methods:
      • Importanced sampling
      • Reject sampling
      • Monte-Carlo Markov Chain (MCMC) sampling
    • Approximation inference:
      • Variational inference
        • Mean-field
      • Expectation propagation

EM

Note about Inverse Problem

less than 1 minute read

Published:

This is a literature review about electromagnetic source imaging (ESI) inverse problem.