OPTIMIZATION
Course objectives
To provide methodologies and algorithmic tools for the formulation, analysis and solution of several optimization problems, with particular reference to those concerning artificial intelligence. The mathematical programming topics and the optimization algorithms characterizing the course will be useful to deeply understand the mathematical models and optimization methods underlying artificial intelligence tools. The students of the course will be able to design and implement algorithms for complex optimization problems and, in particular, for machine learning.
Channel 1
MARCO SCIANDRONE
Lecturers' profile
Program - Frequency - Exams
Course program
Optimization models
Convexity of sets and functions
Linear programming
Integer Linear Programming
Unconstrained optimizazion:
existence of optimal solutions
optimality conditions
unconstrained optimization algorithms
global convergence conditions
line search techniques
gradient methods
stochastic gradient methods
Prerequisites
Linear algebra
Books
Introduction to methods for nonlinear optimization
Luigi Grippo, Marco Sciandrone, Springer, 2023
Lecture notes
Frequency
Optional
Exam mode
Written test with theoretical questions and exercises
Lesson mode
Lectures in classroom
- Lesson code1017662
- Academic year2024/2025
- CourseMathematical Sciences for Artificial Intelligence
- CurriculumSingle curriculum
- Year2nd year
- Semester2nd semester
- SSDMAT/09
- CFU6
- Subject areaFormazione Modellistico-Applicativa