Diverse Randomized Agents Vote to Win

We investigate the power of voting among diverse, randomized software agents. With teams of computer Go agents in mind, we develop a novel theoretical model of two-stage noisy voting that builds on recent work in machine learning. This model allows us to reason about a collection of agents with different biases (determined by the first-stage noise models), which, furthermore, apply randomized algorithms to evaluate alternatives and produce votes (captured by the second-stage noise models). We analytically demonstrate that a uniform team, consisting of multiple instances of any single agent, must make a significant number of mistakes, whereas a diverse team converges to perfection as the number of agents grows. Our experiments, which pit teams of computer Go agents against strong agents, provide evidence for the effectiveness of voting when agents are diverse.

[1]  Craig Boutilier,et al.  Learning Mallows Models with Pairwise Preferences , 2011, ICML.

[2]  Tuomas Sandholm,et al.  The State of Solving Large Incomplete-Information Games, and Application to Poker , 2010, AI Mag..

[3]  Martin Müller,et al.  Fuego—An Open-Source Framework for Board Games and Go Engine Based on Monte Carlo Tree Search , 2010, IEEE Transactions on Computational Intelligence and AI in Games.

[4]  Simon M. Lucas,et al.  A Survey of Monte Carlo Tree Search Methods , 2012, IEEE Transactions on Computational Intelligence and AI in Games.

[5]  Ariel D. Procaccia,et al.  When do noisy votes reveal the truth? , 2013, EC '13.

[6]  L. Thurstone.  A law of comparative judgment. , 1994 .

[7]  Yann Braouezec,et al.  Committee, Expert Advice, and the Weighted Majority Algorithm: An Application to the Pricing Decision of a Monopolist , 2010 .

[8]  Marco LiCalzi,et al.  The Power of Diversity Over Large Solution Spaces , 2011, Manag. Sci..

[9]  Leandro Soriano Marcolino,et al.  Give a Hard Problem to a Diverse Team: Exploring Large Action Spaces , 2014, AAAI.

[10]  Ariel D. Procaccia,et al.  Dynamic Social Choice with Evolving Preferences , 2013, AAAI.

[11]  Tie-Yan Liu,et al.  Learning to Rank for Information Retrieval , 2011 .

[12]  Nicolas de Condorcet.  Essai Sur L'Application de L'Analyse a la Probabilite Des Decisions Rendues a la Pluralite Des Voix , 2009 .

[13]  F. Mosteller.  Remarks on the method of paired comparisons: I. The least squares solution assuming equal standard deviations and equal correlations , 1951 .

[14]  Leandro Soriano Marcolino,et al.  Multi-Agent Team Formation: Diversity Beats Strength? , 2013, IJCAI.

[15]  Lu Hong,et al.  Groups of diverse problem solvers can outperform groups of high-ability problem solvers. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[16]  Tie-Yan Liu,et al.  Learning to rank for information retrieval , 2009, SIGIR.

[17]  R. Luce,et al.  Individual Choice Behavior: A Theoretical Analysis. , 1960 .

[18]  Craig Boutilier,et al.  Optimal social choice functions: A utilitarian view , 2015, Artif. Intell..

[19]  R. Duncan Luce,et al.  Individual Choice Behavior: A Theoretical Analysis , 1979 .

[20]  C. L. Mallows.  NON-NULL RANKING MODELS. I , 1957 .

[21]  R. Plackett.  The Analysis of Permutations , 1975 .

[22]  Scott E. Page,et al.  Collective Wisdom: Some Microfoundations of Collective Wisdom , 2012 .

[23]  Petr Baudis,et al.  PACHI: State of the Art Open Source Go Program , 2011, ACG.

[24]  E. Hellinger,et al.  Neue Begründung der Theorie quadratischer Formen von unendlichvielen Veränderlichen. , 1909 .

[25]  David C. Parkes,et al.  Random Utility Theory for Social Choice , 2012, NIPS.

[26]  David C. Parkes,et al.  Computing Parametric Ranking Models via Rank-Breaking , 2014, ICML.

[27]  Ariel D. Procaccia,et al.  A Maximum Likelihood Approach For Selecting Sets of Alternatives , 2012, UAI.

[28]  L. Bottou,et al.  Generalized Method-of-Moments for Rank Aggregation , 2013 .