Computational Game Theory and Mechanism Design
- To provide an understanding of the inefficiency arising from uncontrolled, decentralized resource allocation.
- To provide a foundation for modelling various mechanism design problems together with their algorithmic aspects.
- To provide the tools and paradigms for the design and analysis of efficient algorithms/mechanisms that are robust in environments that involve interactions of selfish agents.
- To review the links and interconnections between algorithms and computational issues with selfish agents.
- Introduction to Network Games: Reminder of game theory fundamentals (with a focus on network games): solution concepts such as Nash equilibria, correlated equilibria. (2 lectures)
- Load balancing games: existence and complexity of equilibria, price of anarchy. (3 lectures)
- Routing games (atomic and non-atomic selfish routing): existence and complexity of equilibria, price of anarchy, price of stability. (5 lectures)
- Introduction to Mechanism Design: Social Choice, Mechanisms with Money, Dominant Strategies, Characterisations of Incentive Compatible Mechanisms, Bayesian-Nash Implementation. (4 lectures)
- Mechanism Design without Money. (3 lectures)
- Combinatorial Auctions (CA): Single-Minded Bidders, Bidding Languages, Iterative Auctions, Winner Determination, Applications. (4 lectures)
- Profit Maximisation in Mechanism Design. (3 lectures)
- Online Mechanisms (2 Lectures)
- Current Topics in Algorithmic Game Theory (Network Creation Games, Sponsored Search Auctions, Price of Anarchy in Auctions) (4 lectures)
Reading lists are managed at readinglists.liverpool.ac.uk. Click here to access the reading lists for this module.
Explanation of Reading List:
- Have a systematic understanding of current problems and important concepts in the field of computational game theory.
- Ability to quantify the inefficiency of equilibria.
- The ability to formulate mechanism design models or network games for the purpose of modeling particular applications.
- The ability to use, describe and explain appropriate algorithmic paradigms and techniques in context of a particular game-theoretic or mechanism design problem.