| Uitgebreide vaknaam |
Data-driven Optimization |
| Leerdoelen |
After succesful completion of the course, the student is able to:- understand various data-driven optimization models and apply preliminary solution methods to solve them for relevant applications
- explain and apply basics on probability and optimization
- explain and implement various data-driven optimization methods
- explore advanced data-driven optimization methods
- derive parts of theoretical guarantees of algorithms
|
| Omschrijving |
The objective of the course is to understand the fundamentals and relevant applications of data-driven optimization. Particular focus is on machine learning and control theory related fields. The tentative list of topics is:
- Linear and logistic regression
- Support vector machines
- Preliminary convex analysis
- Gradient and stochastic gradient descent
- Accelerated methods
- Online optimization
- Neural networks
- Sample average approximation
- Time permitting: ADMM; Kernel methods
|
| Uren per week |
variabel |
| Onderwijsvorm |
Hoorcollege (LC), Werkcollege (T)
|
| Toetsvorm |
Opdracht (AST), Schriftelijk tentamen (WE)
(Weight of grades: AST 40%, WE 60%)
|
| Vaksoort |
master
|
| Coördinator |
Dr. A.K. Cherukuri
|
| Docent(en) |
Dr. A.K. Cherukuri
|
| Entreevoorwaarden |
The course assumes prior knowledge of Calculus 1 & 2, Linear Algebra, Statistics and Stochastics from the BSc IEM curriculum and Introduction to Optimization from the MSc IEM curriculum. |
| Opmerkingen |
This course was previously registered with course code WMME19011 |
| Opgenomen in |
|
Login
English
Nederlands