EP3301 Spelteori 8,0 hp

Computational Game Theory

The course focuses on areas of game theory that appear to be most relevant for engineering applications. The emphasis is both on theoretical principles and on the application of the theory to problem formulation and problem solving. The course covers a wide range of topics, from different models of non-cooperative games and related equilibrium concepts, to cooperative games. The course also covers topics in mechanism design.

  • Utbildningsnivå

    Forskarnivå
  • Huvudområde

  • Betygsskala

    P, F

Information för forskarstuderande om när kursen ges

Given in odd years in P2.

Lärandemål

Upon completion of the course, the student should be able to:

-          formalize problems that involve more than one decision making entity in a game theoretical context

-          critically assess the research literature in the area

-          use the game theoretical tools and methods to solve problems

-          contribute to the research frontier in the area

Kursens huvudsakliga innehåll

Strategic games in normal form, Kakutani’s fixed point theorem, best reply, Nash equilibrium, dominance solvability, rationalizability, existence of equilibria, zeros-sum games, minimax theorem, cardinality of solutions, refinement of Nash equilibria, robustness of equilibria, Bayesian games, potential games, submodular games, extensive games with perfect information, subgame perfect equilibria, repeated games and folk theorems, stochastic games, Markov perfect equilibria, finite and infinite evolutionary games, replicator dynamic, evolutionary stable states and sets, coalition games, core, kernel, nucleolus, Shapley value, social choice theory, Arrow’s impossibility theorem, implementation in dominant strategies, strategyproof implementation, Gibbard-Sattertwhwaite theorem, implementation with money, Groves mechanism, Clarke’s pivot rule, VCG mechanism, implementation in Nash equilibrium.

Kursupplägg

Lectures, homework problems, presentations on selected topics by the participants, 72 h take home exam

Behörighet

Litteratur

Parts of the course topics are covered in the book
M. J. Osborne, A. Rubinstein, “Course in Game Theory”, MIT Press, Cambridge, Mass., 1994
Alternative: D. Fudenberg, J. Tirole, “Game theory”, MIT Press, Cambridge, Mass., 1991

Lecture notes will be available on the course home page. A list of relevant research and overview articles will be provided.

Examination

  • EXA1 - Examination, 8,0, betygsskala: P, F

Krav för slutbetyg

15 min oral presentation at one of the lectures

75% on weekly home-work problems and the presentation

50% on 72 h take home exam

Ges av

EECS/Nätverk och systemteknik

Kontaktperson

György Dan

Examinator

György Dán <gyuri@kth.se>

Versionsinformation

Kursplan gäller från och med HT2012.
Examinationsinformation gäller från och med VT2019.