(3 units). Bayesian networks, factor graphs, Markov random fields, maximum a posteriori probability (MAP) and maximum likelihood (ML) principles, elimination algorithm, sum-product algorithm, decomposable and non-decomposable models, junction tree algorithm, completely observed models, iterative proportional fitting algorithm, expectation-maximization (EM) algorithm, iterative conditional modes algorithm, variational methods, applications. Courses ELG 5131, ELG 7177 (EACJ 5605) cannot be combined for units. This course is equivalent to EACJ 5131 at Carleton University. Course Component: Lecture
(3 units). Bayesian networks, factor graphs, Markov random fields, maximum a posteriori probability (MAP) and maximum likelihood (ML) principles, elimination algorithm, sum-product algorithm, decomposable and non-decomposable models, junction tree algorithm, completely observed models, iterative proportional fitting algorithm, expectation-maximization (EM) algorithm, iterative conditional modes algorithm, variational methods, applications. Courses ELG 5131, ELG 7177 (EACJ 5605) cannot be combined for units. This course is equivalent to EACJ 5131 at Carleton University. Course Component: Lecture
