DOCSLIB.ORG

The Pennsylvania State University TOPICS in LEARNING AND

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
The Pennsylvania State University TOPICS in LEARNING AND The Pennsylvania State University TOPICS IN LEARNING AND INFORMATION DYNAMICS IN GAME THEORY A Dissertation in Mathematics by Matthew Young © 2020 Matthew Young Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy August 2020 The dissertation of Matthew Young was reviewed and approved by the following: Andrew Belmonte Professor of Mathematics Dissertation Advisor, Chair of Committee Christopher Griffin Professor of Operations Research Jan Reimann Professor of Mathematics Sergei Tabachnikov Professor of Mathematics Syed Nageeb Ali Professor of Economics Alexei Novikov Professor of Mathematics Chair of Graduate Program ii Abstract We discuss the role of learning and information in game theory, and design and investigate four models using different learning mechanisms. The first consists of rational players, one of which has the option of purchasing information about the other player’s strategy before choosing their own. The second is an agent-based public goods model where players incrementally adjust their contribution levels after each game they play. The third is an agent-based rock, paper, scissors model where players choose their strategies by flipping cards with a win-stay lose-shift strategy. The fourth is a machine learning model where we train adversarial neural networks to play arbitrary 2 × 2 games together. We study various aspects of each of these models and how the different learning dynamics in them influence their behavior. iii Table of Contents List of Figures vii List of Tables xi Acknowledgments xii Chapter 1 The Role of Learning in Game Theory 1 1.1 Introduction . 1 1.2 Rational learning after purchasing information . 3 1.3 Agent-based models with simple algorithm learning rules . 4 1.4 Machine learning by adversarial neural networks . 5 Chapter 2 Oracle Games 8 2.1 Introduction . 8 2.2 Preliminary Considerations . 14 2.2.1 Motivating Examples . 15 2.2.2 Definitions . 22 2.3 Fundamental Properties of Oracle Games . 23 2.4 Main Results . 26 2.5 Harmful Information . 34 2.6 Helpful Information . 37 2.7 Multiple Equilibria . 38 2.8 Discussion . 39 Chapter 3 Fair Contribution in a Nonlinear Stochastic Public Goods Model 42 3.1 Introduction . 42 3.1.1 Public Goods Games . 42 3.1.2 Cooperative Behavior in Biology . 44 3.1.3 Modifications of Public Goods Models . 45 3.1.4 Fairness . 47 3.1.5 Our model . 49 iv 3.2 Definition of the Public Goods Game . 49 3.3 Population Dynamics . 51 3.4 Numerical Simulations . 56 3.5 Dynamics in the Presence of a Permanent Freeloader . 68 3.5.1 Multiple Permanent Freeloaders . 70 3.6 Discussion . 72 Chapter 4 Population dynamics in a rock paper scissors model with restricted strategy transitions 74 4.1 Introduction . 74 4.1.1 Win-Stay Lose-Shift . 74 4.1.2 A Biological Basis for a Restriction to Two Strategies . 76 4.1.3 A Biological Basis for Win-Stay Lose-Shift . 77 4.1.4 Our Model . 79 4.2 Discrete Model . 80 4.3 Extinction in a Restricted Transition Population . 83 4.4 Continuous Models . 85 4.5 Discussion . 92 Chapter 5 Neural Networks Playing Games 95 5.1 Introduction . 95 5.1.1 Machine Learning . 95 5.1.2 Game theory and adversarial networks . 97 5.1.3 Our model . 99 5.2 The Model . 99 5.2.1 Construction . 99 5.2.2 Errors . 102 5.2.3 Temptation Games . 108 5.3 Pruning . 111 5.4 Network Comparisons . 114 5.4.1 Euclidean Distance in Weight Space . 114 5.4.2 Polar Projection . 118 5.4.3 Paths Between Networks . 121 5.4.4 Correlation metrics . 124 5.5 Level-k hierarchy in networks . 128 5.6 Discussion . 133 5.6.1 Summary . 133 5.6.2 Extensions of this model . 134 5.6.3 Applications . 135 v Chapter 6 Conclusion 137 6.1 Discussion . 137 6.1.1 Types of Information . 137 6.1.2 Benefit to players . 139 6.1.3 Convergence to a Nash equilibrium . 141 6.1.4 Intelligence . 142 6.2 Future Research . 143 Appendix Master Equation to Langevin Derivation from Chapter 4 146 Bibliography 150 vi List of Figures 2.1 Matching Pennies Extensive form game . 15 2.2 Game tree for the motivating construction of Oracle Games . 16 2.3 Game tree for the standard construction of Oracle Games . 17 √ 2.4 Payment√ in equilibrium√ shown for oracle functions I(x) = (a) x + 1 − 1; (b) x; (c) 2 x. ............................... 19 2.5 Amount paid by Player A at the equilibrium k (above),√ and the resulting response rate I (below) as functions of k when I(x) = kx. 20 2.6 Illustration of the construction to prove Proposition 2 (see text): (top) original given oracle function I(x); (middle) nondecreasing equivalent or- acle function J1(x); (bottom) final nondecreasing, concave oracle function J(x). ..................................... 25 2.7 Harmful Information Extensive form game . 35 √ 2.8 Ea as a function of k when I = kx .................... 37 √ 3.1 Graph of the benefit function b(C) = 400C, with equilibrium and socially optimal values marked in the case where m = 2. 51 3.2 Examples of the Nash equilibrium value Ce plotted as a function of α, for three different return values R = 0.4 (bottom), 0.7 (middle), 0.95 (top). 52 3.3 Representation of a single round of play. 53 3.4 Simulation with m = n = 4 .......................... 55 vii 3.5 Numerical simulation of the model with n = 100 players and group size m = 10, showing the distribution of contributions C of each player around the fairpoint f: t = 0 (top), t = 800 (middle), t = 1600 (bottom). 56 3.6 Average contribution cavg for population of n = 100 players over time, shown for: m = 10 (top), m = 50 (middle), m = 100 (bottom). 57 3.7 Examples of five individual players’ ci over time, in a population of n = 50 players and group size m = 10......................... 58 3.8 E1 behavior . 58 3.9 E1 over time shown for three different initial conditions with n = 100, each with different subgroup sizes: m = 10, 50, 80. 59 3.10 The initial states at t= 0 (left) and at t = 1600 for each group size (right) from the simulation used in Fig. 3.9. 59 3.11 Average decrease rate for E1 as a function of m. 60 3.12 The average value of ci for populations of n = 100 players with one permanent freeloader, as a function of group size m. Each population was numerically simulated and measured ten times at regular intervals between√ t = 200,000 and 300,000. We fixed d = 10,000, and for each m let ui = 40000mC − ci, which gives f = 10, 000 even when m changes. 68 3.13 The average value of ci for populations of n = 100 players with two permanent freeloaders, as a function of m. In this case, two transitions are observed (see text). 71 4.1 Illustration of the population dynamics for one time step . 81 4.2 Trajectory of one simulation for N = 8 . 84 4.3 The average time for strategies played to reach a monoculture (average extinction time) as a function of the total number of cards N, with A = 1/2,B = 1/2,C = 0............................ 85 4.4 Heat map of the discrete system on a grid for (a) N = 40; (b) N = 200. 86 4.5 Trajectory for the deterministic system, with equally spaced initial condi- tions around the plane, for A = 1/2...................... 87 viii 4.6 Parameterized curve of the x and y coordinates of the equilibrium point as functions of A, with A = 0 at (0, 0), and A = 1 at (1, 0) . 87 4.7 Individual system trajectories for total population size N = 800 cards, with the same initial conditions comparing (a) the discrete stochastic system and (b) the Fokker-Planck system. 90 4.8 Heat maps for the discrete and continuous systems with N = 40 after 1,000,000 steps . 91 4.9 Heat map of the difference in the discrete system and continuous system 91 4.10 Extinction Times of discrete system and Fokker-Planck system on a log scale plot, as well as lines of best fit. 92 4.11 Trajectory for Fokker-Planck system with N = 8 . 93 5.1 Network Playing Prisoner Dilemma. Blue lines are positive weights . 102 5.2 Error of a network over time . 103 5.3 Coordination behavior over time . 105 5.4 Error (y) as a function of number of hidden layers (x) . 107 5.5 Error (y) as a function of neurons per hidden layer (x) . 108 5.6 Probability A of choosing strategy sA as a function of x in temptation game 109 5.7 Probability A of choosing strategy sA as a function of x in temptation game after special training . 110 5.8 Log-Log plot of error (y) as a function of neurons pruned (x) . 112 5.9 Histogram of pairwise Euclidean distances for 10 networks . 116 5.10 Error for network pr as a function of r . 117 5.11 Examples of paths in R2 (left) and their corresponding polar projections (right). 119 5.12 Projection of a network path created via backpropagation (left), and a random walk (right) . 120 ix 5.13 Polar projection of three network paths, with θ0 = 0, and paths colored based on the error of the adjacent networks . 123 5.14 Errors of network paths from Figure 5.13 . ..
Load more

Recommended publications
  • Interim Report IR-03-078 Evolutionary Game Dynamics
    International Institute for Tel: 43 2236 807 342 Applied Systems Analysis Fax: 43 2236 71313 Schlossplatz 1 E-mail: [email protected] A-2361 Laxenburg, Austria Web: www.iiasa.ac.at Interim Report IR-03-078 Evolutionary Game Dynamics Josef Hofbauer ([email protected]) Karl Sigmund ([email protected]) Approved by Ulf Dieckmann ([email protected]) Project Leader, Adaptive Dynamics Network December 2003 Interim Reports on work of the International Institute for Applied Systems Analysis receive only limited review. Views or opinions expressed herein do not necessarily represent those of the Institute, its National Member Organizations, or other organizations supporting the work. IIASA STUDIES IN ADAPTIVE DYNAMICS NO. 76 The Adaptive Dynamics Network at IIASA fosters the develop- ment of new mathematical and conceptual techniques for under- standing the evolution of complex adaptive systems. Focusing on these long-term implications of adaptive processes in systems of limited growth, the Adaptive Dynamics Network brings together scientists and institutions from around the world with IIASA acting as the central node. Scientific progress within the network is collected in the IIASA ADN Studies in Adaptive Dynamics series. No. 1 Metz JAJ, Geritz SAH, Meszéna G, Jacobs FJA, van No. 11 Geritz SAH, Metz JAJ, Kisdi É, Meszéna G: The Dy- Heerwaarden JS: Adaptive Dynamics: A Geometrical Study namics of Adaptation and Evolutionary Branching. IIASA of the Consequences of Nearly Faithful Reproduction. IIASA Working Paper WP-96-077 (1996). Physical Review Letters Working Paper WP-95-099 (1995). van Strien SJ, Verduyn 78:2024-2027 (1997). Lunel SM (eds): Stochastic and Spatial Structures of Dynami- cal Systems, Proceedings of the Royal Dutch Academy of Sci- No.
    [Show full text]
  • The Logic of Costly Punishment Reversed: Expropriation of Free-Riders and Outsiders∗,†
    The logic of costly punishment reversed: expropriation of free-riders and outsiders∗ ,† David Hugh-Jones Carlo Perroni University of East Anglia University of Warwick and CAGE January 5, 2017 Abstract Current literature views the punishment of free-riders as an under-supplied pub- lic good, carried out by individuals at a cost to themselves. It need not be so: often, free-riders’ property can be forcibly appropriated by a coordinated group. This power makes punishment profitable, but it can also be abused. It is easier to contain abuses, and focus group punishment on free-riders, in societies where coordinated expropriation is harder. Our theory explains why public goods are un- dersupplied in heterogenous communities: because groups target minorities instead of free-riders. In our laboratory experiment, outcomes were more efficient when coordination was more difficult, while outgroup members were targeted more than ingroup members, and reacted differently to punishment. KEY WORDS: Cooperation, costly punishment, group coercion, heterogeneity JEL CLASSIFICATION: H1, H4, N4, D02 ∗We are grateful to CAGE and NIBS grant ES/K002201/1 for financial support. We would like to thank Mark Harrison, Francesco Guala, Diego Gambetta, David Skarbek, participants at conferences and presentations including EPCS, IMEBESS, SAET and ESA, and two anonymous reviewers for their comments. †Comments and correspondence should be addressed to David Hugh-Jones, School of Economics, University of East Anglia, [email protected] 1 Introduction Deterring free-riding is a central element of social order. Most students of collective action believe that punishing free-riders is costly to the punisher, but benefits the community as a whole.
    [Show full text]
  • 2.4 Mixed-Strategy Nash Equilibrium
    Game Theoretic Analysis and Agent-Based Simulation of Electricity Markets by Teruo Ono Submitted to the Department of Electrical Engineering and Computer Science in partial fulfillment of the requirements for the degree of Master of Science in Electrical Engineering and Computer Science at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY May 2005 ( Massachusetts Institute of Technology 2005. All rights reserved. Author............................. ......--.. ., ............... Department of Electrical Engineering and Computer Science May 6, 2005 Certifiedby.................... ......... T...... ... .·..·.........· George C. Verghese Professor Thesis Supervisor ... / ., I·; Accepted by ( .............. ........1.-..... Q .--. .. ....-.. .......... .. Arthur C. Smith Chairman, Department Committee on Graduate Students MASSACHUSETTS INSTilRE OF TECHNOLOGY . I ARCHIVES I OCT 21 2005 1 LIBRARIES Game Theoretic Analysis and Agent-Based Simulation of Electricity Markets by Teruo Ono Submitted to the Department of Electrical Engineering and Computer Science on May 6, 2005, in partial fulfillment of the requirements for the degree of Master of Science in Electrical Engineering and Computer Science Abstract In power system analysis, uncertainties in the supplier side are often difficult to esti- mate and have a substantial impact on the result of the analysis. This thesis includes preliminary work to approach the difficulties. In the first part, a specific electricity auction mechanism based on a Japanese power market is investigated analytically from several game theoretic viewpoints. In the second part, electricity auctions are simulated using agent-based modeling. Thesis Supervisor: George C. Verghese Title: Professor 3 Acknowledgments First of all, I would like to thank to my research advisor Prof. George C. Verghese. This thesis cannot be done without his suggestion, guidance and encouragement. I really appreciate his dedicating time for me while dealing with important works as a professor and an educational officer.
    [Show full text]
  • Recency, Records and Recaps: Learning and Non-Equilibrium Behavior in a Simple Decision Problem*
    Recency, Records and Recaps: Learning and Non-equilibrium Behavior in a Simple Decision Problem* DREW FUDENBERG, Harvard University ALEXANDER PEYSAKHOVICH, Facebook Nash equilibrium takes optimization as a primitive, but suboptimal behavior can persist in simple stochastic decision problems. This has motivated the development of other equilibrium concepts such as cursed equilibrium and behavioral equilibrium. We experimentally study a simple adverse selection (or “lemons”) problem and find that learning models that heavily discount past information (i.e. display recency bias) explain patterns of behavior better than Nash, cursed or behavioral equilibrium. Providing counterfactual information or a record of past outcomes does little to aid convergence to optimal strategies, but providing sample averages (“recaps”) gets individuals most of the way to optimality. Thus recency effects are not solely due to limited memory but stem from some other form of cognitive constraints. Our results show the importance of going beyond static optimization and incorporating features of human learning into economic models. Categories & Subject Descriptors: J.4 [Social and Behavioral Sciences] Economics Author Keywords & Phrases: Learning, behavioral economics, recency, equilibrium concepts 1. INTRODUCTION Understanding when repeat experience can lead individuals to optimal behavior is crucial for the success of game theory and behavioral economics. Equilibrium analysis assumes all individuals choose optimal strategies while much research in behavioral economics
    [Show full text]
  • Public Goods Agreements with Other-Regarding Preferences
    NBER WORKING PAPER SERIES PUBLIC GOODS AGREEMENTS WITH OTHER-REGARDING PREFERENCES Charles D. Kolstad Working Paper 17017 http://www.nber.org/papers/w17017 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 May 2011 Department of Economics and Bren School, University of California, Santa Barbara; Resources for the Future; and NBER. Comments from Werner Güth, Kaj Thomsson and Philipp Wichardt and discussions with Gary Charness and Michael Finus have been appreciated. Outstanding research assistance from Trevor O’Grady and Adam Wright is gratefully acknowledged. Funding from the University of California Center for Energy and Environmental Economics (UCE3) is also acknowledged and appreciated. The views expressed herein are those of the author and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer- reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications. © 2011 by Charles D. Kolstad. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. Public Goods Agreements with Other-Regarding Preferences Charles D. Kolstad NBER Working Paper No. 17017 May 2011, Revised June 2012 JEL No. D03,H4,H41,Q5 ABSTRACT Why cooperation occurs when noncooperation appears to be individually rational has been an issue in economics for at least a half century. In the 1960’s and 1970’s the context was cooperation in the prisoner’s dilemma game; in the 1980’s concern shifted to voluntary provision of public goods; in the 1990’s, the literature on coalition formation for public goods provision emerged, in the context of coalitions to provide transboundary pollution abatement.
    [Show full text]
  • The Replicator Dynamics with Players and Population Structure Matthijs Van Veelen
    The replicator dynamics with players and population structure Matthijs van Veelen To cite this version: Matthijs van Veelen. The replicator dynamics with players and population structure. Journal of Theoretical Biology, Elsevier, 2011, 276 (1), pp.78. 10.1016/j.jtbi.2011.01.044. hal-00682408 HAL Id: hal-00682408 https://hal.archives-ouvertes.fr/hal-00682408 Submitted on 26 Mar 2012 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Author’s Accepted Manuscript The replicator dynamics with n players and population structure Matthijs van Veelen PII: S0022-5193(11)00070-1 DOI: doi:10.1016/j.jtbi.2011.01.044 Reference: YJTBI6356 To appear in: Journal of Theoretical Biology www.elsevier.com/locate/yjtbi Received date: 30 August 2010 Revised date: 27 January 2011 Accepted date: 27 January 2011 Cite this article as: Matthijs van Veelen, The replicator dynamics with n players and population structure, Journal of Theoretical Biology, doi:10.1016/j.jtbi.2011.01.044 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript.
    [Show full text]
  • The Impact of Beliefs on Effort in Telecommuting Teams
    The impact of beliefs on effort in telecommuting teams E. Glenn Dutcher, Krista Jabs Saral Working Papers in Economics and Statistics 2012-22 University of Innsbruck http://eeecon.uibk.ac.at/ University of Innsbruck Working Papers in Economics and Statistics The series is jointly edited and published by - Department of Economics - Department of Public Finance - Department of Statistics Contact Address: University of Innsbruck Department of Public Finance Universitaetsstrasse 15 A-6020 Innsbruck Austria Tel: + 43 512 507 7171 Fax: + 43 512 507 2970 E-mail: [email protected] The most recent version of all working papers can be downloaded at http://eeecon.uibk.ac.at/wopec/ For a list of recent papers see the backpages of this paper. The Impact of Beliefs on E¤ort in Telecommuting Teams E. Glenn Dutchery Krista Jabs Saralz February 2014 Abstract The use of telecommuting policies remains controversial for many employers because of the perceived opportunity for shirking outside of the traditional workplace; a problem that is potentially exacerbated if employees work in teams. Using a controlled experiment, where individuals work in teams with varying numbers of telecommuters, we test how telecommut- ing a¤ects the e¤ort choice of workers. We …nd that di¤erences in productivity within the team do not result from shirking by telecommuters; rather, changes in e¤ort result from an individual’s belief about the productivity of their teammates. In line with stereotypes, a high proportion of both telecommuting and non-telecommuting participants believed their telecommuting partners were less productive. Consequently, lower expectations of partner productivity resulted in lower e¤ort when individuals were partnered with telecommuters.
    [Show full text]
  • The Replicator Equation on Graphs
    The Replicator Equation on Graphs The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citation Ohtsuki Hisashi, Martin A. Nowak. 2006. The replicator equation on graphs. Journal of Theoretical Biology 243(1): 86-97. Published Version doi:10.1016/j.jtbi.2006.06.004 Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:4063696 Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#LAA The Replicator Equation on Graphs Hisashi Ohtsuki1;2;¤ & Martin A. Nowak2;3 1Department of Biology, Faculty of Sciences, Kyushu University, Fukuoka 812-8581, Japan 2Program for Evolutionary Dynamics, Harvard University, Cambridge MA 02138, USA 3Department of Organismic and Evolutionary Biology, Department of Mathematics, Harvard University, Cambridge MA 02138, USA ¤The author to whom correspondence should be addressed. ([email protected]) tel: +81-92-642-2638 fax: +81-92-642-2645 Abstract : We study evolutionary games on graphs. Each player is represented by a vertex of the graph. The edges denote who meets whom. A player can use any one of n strategies. Players obtain a payo® from interaction with all their immediate neighbors. We consider three di®erent update rules, called `birth-death', `death-birth' and `imitation'. A fourth update rule, `pairwise comparison', is shown to be equivalent to birth-death updating in our model.
    [Show full text]
  • Noisy Directional Learning and the Logit Equilibrium*
    CSE: AR SJOE 011 Scand. J. of Economics 106(*), *–*, 2004 DOI: 10.1111/j.1467-9442.2004.000376.x Noisy Directional Learning and the Logit Equilibrium* Simon P. Anderson University of Virginia, Charlottesville, VA 22903-3328, USA [email protected] Jacob K. Goeree California Institute of Technology, Pasadena, CA 91125, USA [email protected] Charles A. Holt University of Virginia, Charlottesville, VA 22903-3328, USA [email protected] PROOF Abstract We specify a dynamic model in which agents adjust their decisions toward higher payoffs, subject to normal error. This process generates a probability distribution of players’ decisions that evolves over time according to the Fokker–Planck equation. The dynamic process is stable for all potential games, a class of payoff structures that includes several widely studied games. In equilibrium, the distributions that determine expected payoffs correspond to the distributions that arise from the logit function applied to those expected payoffs. This ‘‘logit equilibrium’’ forms a stochastic generalization of the Nash equilibrium and provides a possible explanation of anomalous laboratory data. ECTED Keywords: Bounded rationality; noisy directional learning; Fokker–Planck equation; potential games; logit equilibrium JEL classification: C62; C73 I. Introduction Small errors and shocks may have offsetting effects in some economic con- texts, in which case there is not much to be gained from an explicit analysis of stochasticNCORR elements. In other contexts, a small amount of randomness can * We gratefully acknowledge financial support from the National Science Foundation (SBR- 9818683U and SBR-0094800), the Alfred P. Sloan Foundation and the Dutch National Science Foundation (NWO-VICI 453.03.606).
    [Show full text]
  • Thesis This Thesis Must Be Used in Accordance with the Provisions of the Copyright Act 1968
    COPYRIGHT AND USE OF THIS THESIS This thesis must be used in accordance with the provisions of the Copyright Act 1968. Reproduction of material protected by copyright may be an infringement of copyright and copyright owners may be entitled to take legal action against persons who infringe their copyright. Section 51 (2) of the Copyright Act permits an authorized officer of a university library or archives to provide a copy (by communication or otherwise) of an unpublished thesis kept in the library or archives, to a person who satisfies the authorized officer that he or she requires the reproduction for the purposes of research or study. The Copyright Act grants the creator of a work a number of moral rights, specifically the right of attribution, the right against false attribution and the right of integrity. You may infringe the author’s moral rights if you: - fail to acknowledge the author of this thesis if you quote sections from the work - attribute this thesis to another author - subject this thesis to derogatory treatment which may prejudice the author’s reputation For further information contact the University’s Director of Copyright Services sydney.edu.au/copyright The influence of topology and information diffusion on networked game dynamics Dharshana Kasthurirathna A thesis submitted in fulfillment of the requirements for the degree of Doctor of Philosophy Complex Systems Research Group Faculty of Engineering and IT The University of Sydney March 2016 Declaration I hereby declare that this submission is my own work and that, to the best of my knowledge and belief, it contains no material previously published or written by another person nor material which to a substantial extent has been accepted for the award of any other degree or diploma of the University or other institute of higher learning, except where due acknowledgement has been made in the text.
    [Show full text]
  • 4 Social Interactions
    Beta September 2015 version 4 SOCIAL INTERACTIONS Les Joueurs de Carte, Paul Cézanne, 1892-95, Courtauld Institute of Art A COMBINATION OF SELF-INTEREST, A REGARD FOR THE WELLBEING OF OTHERS, AND APPROPRIATE INSTITUTIONS CAN YIELD DESIRABLE SOCIAL OUTCOMES WHEN PEOPLE INTERACT • Game theory is a way of understanding how people interact based on the constraints that limit their actions, their motives and their beliefs about what others will do • Experiments and other evidence show that self-interest, a concern for others and a preference for fairness are all important motives explaining how people interact • In most interactions there is some conflict of interest between people, and also some opportunity for mutual gain • The pursuit of self-interest can lead either to results that are considered good by all participants, or sometimes to outcomes that none of those concerned would prefer • Self-interest can be harnessed for the general good in markets by governments limiting the actions that people are free to take, and by one’s peers imposing punishments on actions that lead to bad outcomes • A concern for others and for fairness allows us to internalise the effects of our actions on others, and so can contribute to good social outcomes See www.core-econ.org for the full interactive version of The Economy by The CORE Project. Guide yourself through key concepts with clickable figures, test your understanding with multiple choice questions, look up key terms in the glossary, read full mathematical derivations in the Leibniz supplements, watch economists explain their work in Economists in Action – and much more.
    [Show full text]
  • Section 3-Dynamic Gamesand Intellectual Property
    ECON 522 - SECTION 3-DYNAMIC GAMES AND INTELLECTUAL PROPERTY I Dynamic Games Dynamic games are just like static games, except players move sequentially rather than simulta- neously. The definition of Nash equilibrium is still the same (everyone is playing a best response), but now we can end up with NE that don’t make a lot of sense. In the example from class, a start up firm has to choose “enter” or “do not enter,” and the incumbent firm must choose to “fight” or “accommodate.” However, we end up with a strange NE in which the new firm chooses not to enter and the incumbent chooses fight, even though the first firm should figure that a rational incumbent firm would never actually fight should the first firm enter the market. Thus we intro- duced the idea of sequential rationality: everyone believes everyone will behave rationally at every point in time. This leads to a “refinement” of Nash equilibrium, which we call subgame perfect equilibrium (SPE), which is just a NE in which everyone is behaving sequentially rationally. We solve these games using backwards induction, which just means we start from the end of the game and see what choices people would make if they ever reach those end situations, and we work our way back up to the beginning. You can use this method to solve a lot of strategic interactive games, such as tic-tac-toe, or checkers. I.1 Back to a static game- the Public Goods Game from class p Setup: There are 100 people with identical preferences represented by: u = 10 X − x, where X is the total amount of money (including what that individual contributes) donated for a public park, and x is the amount donated by the individual.
    [Show full text]