Abstract

A long-running difficulty with conventional game theory has been how to modify it to accommodate the bounded rationality of all real-world players. A recurring issue in statistical physics is how best to approximate joint probability distributions with decoupled (and therefore far more tractable) distributions. This paper shows that the same information theoretic mathematical structure, known as Product Distribution (PD) theory, addresses both issues. In this, PD theory not only provides a principled formulation of bounded rationality and a set of new types of mean field theory in statistical physics; it also shows that those topics are fundamentally one and the same. 1

Citations

6517	Elements of Information Theory - Cover, Thomas - 1991
2827	Reinforcement Learning: An Introduction - Sutton, Barto - 1998
1585	Game Theory - Fudenberg, Tirole - 1991
1211	The evolution of cooperation - Axelrod, WD - 1981
1196	A Course in Game Theory - Osborne, Rubinstein - 1994
1134	Reinforcement learning: A survey - Kaelbling, Littman, et al. - 1996
786	Bayesian Theory - Bernardo, Smith - 1994
762	Information Theory, Inference and Learning Algorithms Cambridge - MacKay - 2003
561	The Theory of Learning in Games - Fudenberg, Levine - 1998
480	Algorithmic mechanism design - Nisan, Ronen
448	Information theory and statistical mechanics - Jaynes - 1957
391	Advances in prospect theory: cumulative representation of uncertainty - Tversky, Kahneman - 1992
349	Dynamic Noncooperative Game Theory - Basar, Olsder - 1995
340	ET: Probability Theory: The Logic of Science - Jaynes - 2003
250	Improving elevator performance using reinforcement learning - Crites, Barto - 1996
237	Multiagent reinforcement learning: theoretical framework and an algorithm - Hu, Wellman - 1998
188	Correlated Equilibrium as an Expression of Bayesian Rationality,”Econometrica 55 - Aumann - 1987
147	Coalitions among computationally bounded agents - Sandholm, Lesser - 1997
92	Learning mixed equilibria - Fudenberg, Kreps - 1993
80	An introduction to collective intelligence - Wolpert, Tumer - 1999
69	Bounded complexity justifies cooperation in finitely repated prisoner’s dilemma - Neyman - 1985
63	Optimal payoff functions for members of collectives - Wolpert, Tumer - 2001
50	Using collective intelligence to route internet traffic - Wolpert, Tumer, et al. - 1999
40	Coalition, Cryptography and Stability: Mechanisms for Coalition Formation Task Oriented Domains - Zlotkin, Rosenschein - 1994
29	Handbook of Game Theory with Economic Applications - Aumann, Hart - 1992
27	G: The cavity method at zero temperature - Mezard, Parisi
26	Dynamic fictitious play, dynamic gradient play, and distributed convergence to Nash equilibria - Shamma, Arslan - 2005
22	How can Statistical Mechanics contribute to social science - Durlauf - 1999
21	Complexity in Economic Theory: Inductive Reasoning and Bounded - Arthur
21	AIJ Editorial: Economic Principles of Multi-Agent Systems - Boutilier, Shoham, et al. - 1997
19	Product distribution theory for control of multi-agent systems - Lee, Wolpert - 2004
18	Distributed control by Lagrangian steepest descent - Wolpert, Bieniawski - 2004
14	Adaptive, distributed control of constrained multi-agent systems - Bieniawski, Wolpert
13	Fleet assignment using collective intelligence - Antoine, Bieniawski, et al. - 2004
12	A psychological perspective on economics - Kahneman - 2003
11	Adaptive multi-agent systems for constrained optimization - Macready, Wolpert - 2004
11	Discrete, continuous, and constrained optimization using collectives - Bieniawski, Wolpert, et al. - 2004
9	2004c), What information theory says about best response, binding contracts, and collective intelligence - Wolpert
8	Factoring a canonical ensemble - Wolpert - 2003
8	Product distribution theory and semicoordinate transformations - Airiau, Wolpert
7	Self-dissimilarity as a high dimensional complexity measure - Wolpert, Macready
6	Large nonanonymous repeated games - Al-Najjar, Smorodinsky - 2001
5	Adaptive metropolis hastings sampling using product distributions - Wolpert, Lee - 2004
5	Adaptive distributed control: beyond single-instant categorical variables - Wolpert, Bieniawski
5	Beyond mechanism design - Wolpert, Tumer - 2002
2	intelligence, data routing and braess’ paradox - “Collective
2	Animal Contests as Evolutionary Games - Mesterton-Gibbons, Adams - 1998
2	Factoring a canonical ensemble,” 2003, preprint condmat/0307630 - Wolpert
1	Optimal combinations of imperfect objects,” Phys - Challet, Johnson - 2002
1	Economic history and game theory: A survey,” in Handbook of Game Theory with Economic Applications - Greif - 1999