Files
2x2-Games-Probability-Groups.cpp
726-Games-Brute-Force-Proof.cpp
NC-Agent-Simlation-C__-Code.zip
SimpsonThesis-April13.pdf
SQ-Agent-Simulation-C__-Code.zip
Usage
  • 261 views
  • 479 downloads

Simulating Strategic Rationality

  • Author / Creator
    Simpson, John
  • This project explores the intersection of three topics: games, rationality and simulation. There are four major results produced by this exploration. First, it is argued that whether or not a particular process can count as a simulation for a particular person is entirely dependent on whether or not that person is able to see the process in questions as being relevantly and appropriately similar to the process that it is intended to be a simulation of. In short, whether or not something can count as a simulation is entirely observer relative. Second, a proof is offered that there are exactly 726 meaningfully different 2x2 games. This proof addresses a confusion in the game theory literature regarding exactly how many 2x2 games are possible. Third, an argument based on the statistical likelihood of encountering each of the various 2x2 games is advanced; presenting a serious challenge for the common practice of focusing on only a small number of games such as the Prisoner's Dilemma, Chicken, and Stag Hunt. From a small set of assumptions it follows these attention grabbing games are likely to be encountered at most eight percent of the time. This result has important implications for designing artificial intelligences and for understanding human behaviour. Finally, it is shown that under certain stipulated conditions playing games using the model of rationality presented by traditional economics, decision, and game theory, results in sub optimal outcomes. Specifically, there is at least one case preventing population inhabitants from having full information provides the population that they belong to with adaptive advantages. The populations used to show this behaviour choose from a status quo position and were inspired by the framing effect known as the status quo bias. In addition to these major results it is also shown that simply endowing artificial agents with theory of mind is not necessarily beneficial and an argument is made for creating/maintaining spaces for play within science.

  • Subjects / Keywords
  • Graduation date
    Spring 2010
  • Type of Item
    Thesis
  • Degree
    Doctor of Philosophy
  • DOI
    https://doi.org/10.7939/R3889H
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.
  • Language
    English
  • Institution
    University of Alberta
  • Degree level
    Doctoral
  • Department
  • Supervisor / co-supervisor and their department(s)
  • Examining committee members and their departments
    • Rockwell, Geoffrey (Philosophy and Humanities Computing)
    • MacIntosh, Duncan (Philosophy, Univiersity of Dalhousie)
    • Quamen, Harvey (English & Film Studies and Humanities Computing)
    • Cooper, Wes (Philosophy)