BECM33MLF - Machine Learning Fundamentals

Links: BECM33MLF, Lectures, Labs and Seminars, upload system, course schedule, schedule of academic year.

The goal of statistical machine learning is to design systems that incorporate models and algorithms capable of learning to solve problems from examples and prior knowledge. This course is organized around two main objectives. First, it aims to clarify the fundamental principles of machine learning—such as risk minimization, maximum likelihood estimation, and Bayesian learning—and to explore their theoretical foundations. Second, it focuses on introducing core models for classification and regression, and on demonstrating how these models can be effectively learned by applying these foundational concepts.

• Teachers: Vojtech Franc ,Jakub Paplham

• Prerequisites:
  • elements of probability theory, statistics, linear algebra and optimization

• Course format: (2/2)
  • lectures: weekly, Tuesday 14:30-15:15, KN:A-309
  • seminars: weekly, Tuesday 16:15-17:45, KN:A-309

• Grading/Credits:
  • Credits: 6 CP
  • Thresholds for passing: at least 50% of the regular points in the homeworks and at least 50% of the regular points in the exam
  • Weights for final grading: 40% homeworks + 60% written exam = 100% (+ bonus points, e.g. challenge or finding typos in lecture slides)
  • If thresholds for passing are met, the grade is assigned according to the following table:

• Textbooks and References:
  • [1] Yaser S. Abu-Mostafa et al. Learning From Data – A short course. AMLbook.com. 2012.PDF
  • [2] Shai Shalev-Shwartz et al. Undestanding Machine Learning: From Theory to Algorithms. Cambridge press. 2014. PDF
  • [3] M. Mohri, A. Rostamizadeh and A. Talwalkar, Foundations of Machine Learning, MIT Press, 2012 [PDF]

Lectures

Seminars