Theories and methods for automating and representing knowledge with an emphasis on learning from input/output data. The course covers a wide variety of approaches, including Supervised Learning, Neural Nets and Deep Learning, Reinforcement Learning, Expert Systems, Bayesian Learning, Fuzzy Rules, Genetic Algorithms, and Swarm Intelligence. Each student focuses on two of these approaches and creates a term project. Laboratory course. Prerequisite: MET CS 521 or MET CS 622 or MET CS 673. Or instructor's consent.