Infopaket

Cоnvex optimization methods

ДВС.3.01.03

Вибіркова дисципліна для ОП

Second

2021/2022

4 Semester

4

LO14. Apply innovative approaches in computer science and information technology.

Full-time form

1. Successful mastering of courses: discrete mathematics; linear algebra and analytic geometry; Operations Research; mathematical analysis; functional analysis. 2. Knowledge: basic concepts and methods of mathematical programming; basics of convex analysis; basic information on the theory of second-order curves; theory of boundaries and functions of a real variable.

The aim of the discipline is mastering the knowledge and skills of theory and methods of non-smooth optimization in construction and analysis of algorithms for solving applied optimization problems. As a result of studying the discipline the student must: know: basic concepts of non-smooth optimization methods and conditions of their effective application in applied optimization problems; be able to: formulate mathematical models of applied optimization problems and use subgradient methods to minimize non-smooth convex functions to solve them.

1. Glushkov V.M. Basics of Paperless Informatics. – М.: Nauka, 1982. – 552 p. 2. Shor N.Z. Minimization Methods for Non-Differentiable Function. – Kiev: Naukova Dumka, 1979. – 200 p. 3. Polyak B.T. Introduction to Optimization. – М.: Nauka, 1983. – 384 p. 4. Stetsyuk P.I. Ellipsoid Methods and r-Algorithms. – Chisinau: Evryka, 2014. – 488 p.

Lectures, exam.

- Intermediate assessment: 1. Lection 1-13 Work: LO1.1, LO2.1, LO2.2, LO3.1, LO4.1 — 60 points. - Final assessment (exam): - maximum number of points: 40 points; - learning outcomes to be assessed: LO1.1, LO1.2, LO2.1, LO2.2, LO3.1. - exam form: written. - 3 written assignments (3 theoretical questions). Students that earned less than 36 points are not admitted to exam.

Ukrainian