LFU:online

How can we design learning software systems that adjust their parameters according to example data, continually optimize their own performance, and/or automatically adapt to changing contexts? This course conveys knowledge of basic techniques and competences in the formulation and solution of problems of machine learning.

Fundamentals in statistical methods. Suprevised learning: classification and regression. Unsupervised learning: clustering, density estimation, and dimensionality redection.

The lecture covers theoretical material that the complementary proseminar exercises in discussions, written problems, and programming projects.

Course assessment is based on regular written and/or oral contribution by participants.

This course will take most material from Pattern Recognition and Machine Learning by Chris Bishop.

Additional material will be taken from other sources and will be provided to the students before the lecture.

Basic Python programming skills are requried in this course.

It is recommened to take this course after completion of "VU Introduction to Programming: Programming in Python" from the Minor Digital Science or an equivalent course.

The acceptance procedure is based on prioritised randomisation. Students advanced in completion of the Minor Digital Science get precedence, especialy these who passed module 1 (Python). 

This course is for all bachelor students and master students excluding Computer Science students as the underlying course 703.075 is mandatory in the Bachelor program, and the Master program offers dedicated machine learning courses.

The course is blocked over 9 first weeks of the semester, therefore it is 5 hours weekly!

The lecture part is within 703.075. The content of lectures in both courses is identical up to week 9. In 703.075, additionally Reinforcement Learning is covered.