Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the fundamentals and advanced concepts of Bayesian networks through this comprehensive lecture series from UCLA's Automated Reasoning Group. Master the essential foundations of propositional logic and probability calculus, including beliefs, hard evidence, and soft evidence. Delve into Bayesian network syntax, semantics, independence properties, and d-separation principles. Learn to build effective Bayesian networks through systematic approaches and understand the complexity of probabilistic queries. Discover multiple inference techniques including variable elimination, the jointree algorithm, inference by conditioning, and loopy belief propagation. Examine arithmetic circuits, Sum-Product Networks (SPNs), and Probabilistic Sentential Decision Diagrams (PSDDs) for advanced computational methods. Study parameter learning with both complete and incomplete data, network structure learning algorithms, and Bayesian learning approaches with discrete parameter sets and Dirichlet priors. Investigate causality concepts including interventions and counterfactuals, sensitivity analysis techniques, and modern applications to classifier reasoning and explanation. Gain practical skills in modeling real-world problems using probabilistic graphical models while developing a deep theoretical understanding of uncertainty reasoning in artificial intelligence systems.
