Regression Equilibrium

Prediction is a well-studied machine learning task, and prediction algorithms are core ingredients in online products and services. Despite their centrality in the competition between online companies who offer prediction-based products, the strategic use of prediction algorithms remains unexplored. The goal of this paper is to examine strategic use of prediction algorithms. We introduce a novel game-theoretic setting that is based on the PAC learning framework, where each player (aka a prediction algorithm aimed at competition) seeks to maximize the sum of points for which it produces an accurate prediction and the others do not. We show that algorithms aiming at generalization may wittingly mispredict some points to perform better than others on expectation. We analyze the empirical game, i.e., the game induced on a given sample, prove that it always possesses a pure Nash equilibrium, and show that every better-response learning process converges. Moreover, our learning-theoretic analysis suggests that players can, with high probability, learn an approximate pure Nash equilibrium for the whole population using a small number of samples.

[1]  Yishay Mansour,et al.  Competing Bandits: Learning Under Competition , 2017, ITCS.

[2]  L. Shapley,et al.  Potential Games , 1994 .

[3]  Ariel D. Procaccia,et al.  Collaborative PAC Learning , 2017, NIPS.

[4]  Ariel D. Procaccia,et al.  Algorithms for strategyproof classification , 2012, Artif. Intell..

[5]  I. Glicksberg.  A FURTHER GENERALIZATION OF THE KAKUTANI FIXED POINT THEOREM, WITH APPLICATION TO NASH EQUILIBRIUM POINTS , 1952 .

[6]  Ariel D. Procaccia,et al.  Incentive compatible regression learning , 2008, SODA '08.

[7]  Yakov Babichenko,et al.  Empirical Distribution of Equilibrium Play and Its Testing Application , 2013, Math. Oper. Res..

[8]  Peter L. Bartlett,et al.  Neural Network Learning - Theoretical Foundations , 1999 .

[9]  Ariel D. Procaccia,et al.  Strategyproof Linear Regression in High Dimensions , 2018, EC.

[10]  Yannai A. Gonczarowski,et al.  Efficient empirical revenue maximization in single-parameter auction environments , 2016, STOC.

[11]  Tim Roughgarden,et al.  On the Pseudo-Dimension of Nearly Optimal Auctions , 2015, NIPS.

[12]  Shai Ben-David,et al.  Understanding Machine Learning: From Theory to Algorithms , 2014 .

[13]  Moshe Tennenholtz,et al.  Best Response Regression , 2017, NIPS.

[14]  Vladimir Vapnik,et al.  Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities , 1971 .

[15]  Roger B. Myerson,et al.  Optimal Auction Design , 1981, Math. Oper. Res..

[16]  Aranyak Mehta,et al.  Playing large games using simple strategies , 2003, EC '03.

[17]  Christos H. Papadimitriou,et al.  Strategic Classification , 2015, ITCS.

[18]  Richard Cole,et al.  The sample complexity of revenue maximization , 2014, STOC.

[19]  I. Althöfer.  On sparse approximations to randomized strategies and convex combinations , 1994 .

[20]  David H. Wolpert,et al.  No free lunch theorems for optimization , 1997, IEEE Trans. Evol. Comput..

[21]  I. Erev,et al.  Small feedback‐based decisions and their limited correspondence to description‐based decisions , 2003 .

[22]  H. Simon,et al.  Rational choice and the structure of the environment. , 1956, Psychological review.

[23]  Adam Tauman Kalai,et al.  Dueling algorithms , 2011, STOC '11.

[24]  R. Rosenthal.  A class of games possessing pure-strategy Nash equilibria , 1973 .

[25]  Leslie G. Valiant,et al.  A theory of the learnable , 1984, STOC '84.

[26]  H. Hotelling.  Stability in Competition , 1929 .

[27]  Alvin E. Roth,et al.  A choice prediction competition: Choices from experience and from description , 2010 .

[28]  Norbert Sauer,et al.  On the Density of Families of Sets , 1972, J. Comb. Theory A.