Mathematical foundations of machine learning and forecasting

Learning objectives:

The course focuses on a current area of mathematics and information technology — the mathematical foundations of machine learning and forecasting. The course will cover various machine learning algorithms. It will present the theory of linear classifiers, algorithms for constructing classifiers for linearly separable datasets, the support vector machine method, and its generalization — kernel methods in machine learning. Probabilistic machine learning algorithms will also be presented. In addition, the course will include sections on intelligent forecasting methods. It will cover algorithms for exponential weighting of expert forecasts (both deterministic and probabilistic), V. G. Vovk’s aggregating algorithm, and the classifier boosting algorithm. The main approaches to reinforcement learning will also be presented.

Upon completing the course, students will gain knowledge in the mathematical foundations of machine learning and forecasting. They will be familiarized with mathematical models for machine learning and forecasting and will acquire the skills to apply this knowledge to solve current theoretical and practical problems in machine learning and forecasting.

At the end of the course, a credit test will be held.

To take the course, students need knowledge of mathematical logic, linear algebra, and discrete mathematics.

Classes are held at the Main Building of Lomonosov Moscow State University, Room 16‑10

The course program includes 24 sessions: 12 lectures and 12 practical sessions

Format: offline

Course start date: February 19, 2025

Classes will be held on Wednesdays from 15:00 to 16:30

Enrollment for the classes on the website

Course Telegram channel

Enrollment for the 2025 course is closed