# Графовые модели (Probabilistic Graphical Models)

Probabilistic graphical models (PGMs) such as Bayesian Networks are a popular way to represent and reason about domains involving uncertainty. Given information about some features of the environment, the models can be used to reach conclusions about other unknown facts. The course will teach students about reasoning algorithms on PGMs and various algorithms to automatically construct PGMs from data. Topics covered include Bayesian networks, undirected networks, Gaussian networks, causal models, inference algorithms (e.g. variable elimination, particle-based approximate inference), network learning (e.g. parameter estimation, structure learning)