INVESTIGATING EQUILIBRIA OF GAMES WITH UNCERTAIN UTILITIES
Lead Research Organisation:
CARDIFF UNIVERSITY
Department Name: Sch of Mathematics
Abstract
In 1953 Shapley wrote the first seminal piece of work on stochastic games. In that paper these games were defined as ones in which the "the play proceeds by steps from position to position, according to transition probabilities controlled jointly by the two players".
This PhD research aims to study stochastic games not only in this traditional sense but also by considering strategic games with utilities that are outputs of stochastic models. Examples of this include Markov queuing processes. These games arise when considering strategic behaviour in queuing models.
The project involves a mix of numerical experimentation as well as developing theoretical results and algorithms about/for the computation of equilibria.
This PhD research aims to study stochastic games not only in this traditional sense but also by considering strategic games with utilities that are outputs of stochastic models. Examples of this include Markov queuing processes. These games arise when considering strategic behaviour in queuing models.
The project involves a mix of numerical experimentation as well as developing theoretical results and algorithms about/for the computation of equilibria.
Organisations
People |
ORCID iD |
Vincent Knight (Primary Supervisor) | |
Nikoleta Glynatsi (Student) |
Publications
Glynatsi N
(2018)
An evolutionary game theoretic model of rhino horn devaluation
in Ecological Modelling
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/N509449/1 | 30/09/2016 | 29/09/2021 | |||
1798698 | Studentship | EP/N509449/1 | 30/09/2016 | 30/03/2020 | Nikoleta Glynatsi |
Description | My work explores the well known game of the Prisoner's Dilemma. My key fundings include a presenting compact method of identifying the best response memory- one strategy against a given set of opponents, and evaluates whether it behaves in a zero- determinant way which in turn indicates whether it can be extortionate. This is also done in evolutionary settings. Moreover, it introduces a well designed framework that allows the comparison of an optimal memory one strategy and a more complex strategy which has a larger memory. This is used to identify conditions for which defection is stable; thus identifying environments where cooperation will not occur. |
Exploitation Route | The source code for my work has been made available and it usable by other researchers. More over, all the data sets used in my research are also available. These can be used for further analysis and insights in the field |
Sectors | Other |