All programmes > Course: Optimization: special topics

Optimization: special topics (B-KUL-D0M90B)

6.0 ECTS English 39.0 Second termSecond term Advanced Cannot be taken as part of an examination contract
Spieksma Frederik
POC Engelstalige Masters FEB

Aims

This is an advanced course in optimization. Different topics will be discussed among which: network optimization, linear integer optimization, Lagrange relaxation. We will describe formulations, applications, and methods for optimization problems, with a special emphasis on combinatorial optimization problems.
The goal is to be able to translate a given optimization question into a formulation, and to be able to make a balanced judgement of the different options of solving the given optimization problem.

Previous knowledge

Order of enrolment:
The following course units should be:
* successfully completed: /
* taken before: /
* at least taken at the same time: /
Clarification:
Linear Optimization, Operations Research

Content

We start with formulations of different (combinatorial) optimization problems, and discuss their pros and cons.We see applications in different domains, and we pay attention to different solution methods. In particular, techniques that will be explained are: column generation for solving huge linear optimization problems, branch-and-bound methods, branch-and-price methods (Dantzig-Wolfe decomposition), approximation algorithms, and heuristics (local search methods). We informally discuss computational complexity explaining differences in the difficulty of solving combinatorial problems. There might be some variation in the contents of the course depending upon the interests of the participants.

Course material

Syllabus

Is also included in other courses

Activities

6.0 ects. Optimization: special topics (B-KUL-D0M90a)

6.0 ECTS English 39.0 Second termSecond term
Spieksma Frederik
POC Engelstalige Masters FEB

Content

We start with formulations of different (combinatorial) optimization problems, and discuss their pros and cons.We see applications in different domains, and we pay attention to different solution methods. In particular, techniques that will be explained are: column generation for solving huge linear optimization problems, branch-and-bound methods, branch-and-price methods (Dantzig-Wolfe decomposition), approximation algorithms, and heuristics (local search methods). We informally discuss computational complexity explaining differences in the difficulty of solving combinatorial problems. There might be some variation in the contents of the course depending upon the interests of the participants.

Aims

This is an advanced course in optimization. Different topics will be discussed among which: network optimization, linear integer optimization, Lagrange relaxation. We will describe formulations, applications, and methods for optimization problems, with a special emphasis on combinatorial optimization problems.
The goal is to be able to translate a given optimization question into a formulation, and to be able to make a balanced judgement of the different options of solving the given optimization problem.

Description of learning activities

There is a class once a week. Each week there are exercises to be made, which are discussed in the next class. There is also a single presentation of a scientific paper by each student

Course material

Cursustekst: "Optimization: special topics"

Evaluation

Evaluation : Optimization: special topics (B-KUL-D2M90b)

Mode of evaluation : Written
Category : interim evaluations plus final examination during examination period
Type of evaluation : Open book, Presentation

Explanation

Bepaling examenresultaat 
 
* Het examen wordt beoordeeld door 
de docent, zoals meegedeeld via Toledo en de examenregeling. Het 
resultaat wordt berekend en uitgedrukt met een geheel getal op 20.
 
*
Het examen is open boek. Indien meerkeuzevragen gesteld worden, wordt 
er bij de beoordeling geen giscorrectie toegepast.

* Bij
een onvoldoende op het deel Presentatie kan een correctie naar beneden 
plaatsvinden. De toegepaste correctie staat in verhouding tot de omvang 
van het tekort, en wordt meegedeeld via TOLEDO.
 
Evaluatie
derde examenperiode

* De student heeft per academiejaar tweemaal
de kans deel te nemen aan het examen: een eerste keer in de eerste of 
tweede examenperiode, overeenkomstig het semester waarin het 
opleidingsonderdeel staat geprogrammeerd, en een tweede keer in de derde
examenperiode. 

* De evaluatiekenmerken van de derde 
examenperiode zijn gelijk aan deze van de eerste of tweede 
examenperiode