3 Cr. (Hrs.:3 Lec.)
Introduction to the framework of learning from examples. Topics include various learning algorithms such as neural networks, Bayesian networks, and genetic algorithms, and generic learning principles such as bias/variance, MDL principle, and ethical considerations. Review statistical learning techniques, yet focuses on non-statistical techniques. Students may not take this course for both 400 and 500 level credit. Prerequisite CSCI 332 or consent of instructor. (2nd)
R1. Be comfortable with machine learning fundamentals including probability, linear algebra, data analysis, the overall machine learning process and general dimensions of machine learning problems
R2. Have reviewed the statistical techniques of regression, clustering and the nearest neighbor approach.
R3. Understand and be able to implement machine learning algorithms such as neural networks, Bayesian networks and genetic algorithms.
R4. Be able to discuss tradeoffs between different machine learning algorithms, hyperparameter selection heuristics, and bias/variance.
R5. Understand performance metrics and what measures to use to compare results from different models.
R6. Be able to identify and implement ensemble learning techniques.
R7. Be able to develop workable representations for the various approaches, and identify situations in which data manipulation must be used prior to learning.
R8. Understand ethical considerations and assumptions behind the development of a learned model.
R9. Demonstrate the ability to implement one or more learning techniques using a real-life dataset.