Stochastic Gradient Techniques in Optimization and Learning

2019-2020
Dit vak wordt in het Engels aangeboden. Omschrijvingen kunnen daardoor mogelijk alleen in het Engels worden weergegeven.

Doel vak

In this course the student will learn how to find unbiased gradient
estimators and how to apply them in stochastic simulation-based
optimization and learning algorithms. After successful participation in
this course the student will be able to conduct a gradient-based
stochastic optimization solution to real life problems.

Inhoud vak

In presence of uncertainty, gradients typically fail to be available in
analytical form and optimization has to resort to simulation-based
algorithms. Unbiased gradient estimators are a main ingredient in
simulation-based optimization methods. The focus of this course is on
unbiased gradient estimators and their application in stochastic
simulation-based optimization and learning algorithms. Next to classical
stochastic gradient methods, this course also covers a range of related
topics such as model and parameter insecurity, robust optimization and
sample average approximation. Applications will stem from a wide range
of domains from Financial Engineering over Inventory Management to
Waiting Time Minimization and neural networks.

This is course on advanced simulation techniques. The methodological
part of the course focusses on the theory of recursive learning and
optimization algorithms known as stochastic approximation.

Onderwijsvorm

Combined lectures and tutorials

Toetsvorm

Final exam – Individual assessment
Individual assignment - Individual assessment

Literatuur

Handout of monograph “Gradient based Stochastic Optimization”, B.
Heidergott and F. Vásquez-Abad, 2018.

Overige informatie

The course is suitable to be taken in an exchange program.

Aanbevolen voorkennis

Analysis, basic probability theory, basic programming

Algemene informatie

Vakcode E_EORM_STGOL
Studiepunten 6 EC
Periode P2
Vakniveau 400
Onderwijstaal Engels
Faculteit School of Business and Economics
Vakcoördinator prof. dr. B.F. Heidergott
Examinator prof. dr. B.F. Heidergott
Docenten dr. A.A.N. Ridder
prof. dr. B.F. Heidergott

Praktische informatie

Voor dit vak moet je zelf intekenen.

Voor dit vak kun je last-minute intekenen.

Werkvormen Hoorcollege, Werkgroep
Doelgroepen

Dit vak is ook toegankelijk als: