2023/2024

Computational Learning Theory

Type: Optional course (faculty)

Delivered by: Big Data and Information Retrieval School

When: 1 module

Online hours: 52

Open to: students of one campus

Language: English

ECTS credits: 3

Contact hours: 10

Course Syllabus

Abstract

Our course aims at introducing machine learning and its paradigms. We will provide basic tools and concepts for theoretical justification of reliability of statistical learning including the concentration of measure phenomenon, empirical process theory, VC-dimension, and Rademacher’s complexity. These tools will be then applied to the analysis of popular algorithms in supervised and unsupervised learning. It is supposed that students have a solid background in calculus, linear algebra, and probability theory in order to fluently follow the course.