Inference & Learning

• Monte-Carlo Algorithms

•Direct Sampling by rejection, pebble game, by mapping

•Importance sampling

•Direct brut-force sampling

•Stochastic algorithms for Optimization

• Markov-Chain Monte-Carlo (MCMC)

•Ising model [running example of Graphical Model]

•Gibbs sampling, Metropolis-Hastings. Exact sampling (for special cases, e.g. treeс with link to Dynamic Programming)

• Algorithms for Inference & Learning

•Variational algorithms (mean-field, belief propagation/message passing –  example of Ising model)

•Neural Networks Algorithms – training/learning, executing/inference (primer on tensorflow or pytorch or julia-flux)