Machine learning with reinforcement: regular and deep - course RUB 43,900. from IBS Training Center, training 30 hours, Date: February 12, 2024.

This course is aimed at programmers using Python or working with machine learning who want to understand the capabilities of RL in more detail.
The course focuses on issues that arise when working with RL in real life, and also examines the basic algorithms that underlie the latest RL systems.
Students will gain theoretical knowledge not only about RL, but also about machine learning in general, as well as the opportunity to apply this knowledge in practice. Starting with the basics of linear algebra, calculus and probability theory, we will then look at issues of dynamic programming and Markov processes, common Q-learning methods and its deep variants, as well as some gradient methods.
In this course we aim to provide a comprehensive overview of the topic, at least with respect to the various basic techniques described in the literature.
The second half of the course is devoted exclusively to RL with neural networks, with a detailed discussion of current research and an explanation of the various applications of this method.

In addition, practical tools (from Pytorch to Ray) are discussed that students will be able to use in the exercises.