PristupačnostOptimization for Machine Learning
Data is displayed for the academic year: 2025./2026.
Lecturers
Course Description
This course provides an overview of modern optimization methods for applications in machine learning and data science. In the introductory part of the course, students are introduced to the basic concepts of convex analysis. An overview of convex optimization problems is given, with particular attention to examples of convex problems encountered in machine learning. The first part of the course is devoted to gradient methods. The central part of the course deals with the concept of duality and its applications. The last part of the course deals with selected constrained optimization algorithms (barrier method; penalty method; primal-dual interior point method). Finally, we will deal with Physically-Informed Neural Networks (PINNs) for optimal control of partial differential equations. Program solutions are implemented using the Python program package.
Study Programmes
Postgraduate doctoral study programme
Literature
Guanghui Lan (2021.), First-order and Stochastic Optimization Methods for Machine Learning, Springer
Suvrit Sra, Sebastian Nowozin, Stephen J. Wright (2012.), Optimization for Machine Learning, MIT Press
Charu C. Aggarwal (2020.), Linear Algebra and Optimization for Machine Learning, Springer Nature
Stephen Boyd, Stephen P. Boyd, Lieven Vandenberghe (2004.), Convex Optimization, Cambridge University Press
General
ID 240528
Summer semester
6 ECTS