FDD3402 Kombinatorisk optimering 6,0 hp

The course aims to give a foundation of advanced techniques that lead to efficient exact algorithms. After an introduction to fundamental polyhedral concepts such as integral polyhedra and their connection to totally unimodular matrices, the course focuses on matroids and their connection to greedy algorithms.

The last part of the course introduces expander graphs from a combinatorial optimization point of view.

After the course, the students should

• know basic concepts from polyhedral combinatorics
• be able to recognize several types efficiently solvable problems based on polyhedral techniques and matroids
• be able to understand techniques from combinatorial optimization used in research papers
• have an enhanced base of techniques to approach open algorithmic problems.

The course is selfcontained, but it is beneficial to have basic knowledge of optimization problems and in particular linear programming as it was provided, for instance, in the course DD3390, Approximation Algorithms given by Ola Svensson in 2010.

När kurs inte längre ges har student möjlighet att examineras under ytterligare två läsår.

Examinator beslutar, baserat på rekommendation från KTH:s handläggare av stöd till studenter med funktionsnedsättning, om eventuell anpassad examination för studenter med dokumenterad, varaktig funktionsnedsättning.

Examinator får medge annan examinationsform vid omexamination av enstaka studenter.

Per Austrin

• Vid grupparbete har alla i gruppen ansvar för gruppens arbete.
• Vid examination ska varje student ärligt redovisa hjälp som erhållits och källor som använts.
• Vid muntlig examination ska varje student kunna redogöra för hela uppgiften och hela lösningen.

The course is mainly targeted to graduate students at KTH in computer science and mathematics, but also open for advanced undergraduate students.