Max amount of FITech students: 30
In this course the focus is especially on classifiers and classification. Other topics covered include statistical regression, Bayesian decision theory and feature extraction. Contents of this course are Bayesian decision theory, discriminant functions, parametric and non-parametric classification, feature extraction, classifier design with example classifiers, statistical regression methods.
After completing the course, the student
- can design simple optimal classifiers from the basic theory and assess their performance
- can explain the Bayesian decision theory and apply it to derive minimum error classifiers and minimum cost classifiers
- can apply the basics of gradient search method to design a linear discriminant function
- can apply regression techniques to practical machine learning problems.
Course is based on Duda RO, Hart PE, Stork DG: Pattern classification, John Wiley & Sons Inc., 2nd edition, 2001. There are also additional handouts.
Course contains face-to-face teaching, guided laboratory work and independent assignment. Laboratory work is supervised by assistants who also check that the task assignments are completed properly. The independent task assignment is graded. The course ends with a written exam.
See the schedules for lectures and exercises here (to be updated).
More information in University of Oulu’s WebOodi course page.
You can get a digital badge after completing this course.
tekoäly artificial intelligence AI koneoppiminen luokittelu valitsin valitsimet gradientti
Further information about the studies
Contact person for applications