Nash Convergence of Mean-Based Learning Algorithms in First-Price Auctions

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Deng, Xiaotie, Hu, Xinyan, Lin, Tao, Zheng, Weiqiang
Format: Preprint
Publié: 2021
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912544585678848
author Deng, Xiaotie
Hu, Xinyan
Lin, Tao
Zheng, Weiqiang
author_facet Deng, Xiaotie
Hu, Xinyan
Lin, Tao
Zheng, Weiqiang
contents The convergence properties of learning dynamics in repeated auctions is a timely and important question, with numerous applications in, e.g., online advertising markets. This work focuses on repeated first-price auctions where bidders with fixed values learn to bid using mean-based algorithms -- a large class of online learning algorithms that include popular no-regret algorithms such as Multiplicative Weights Update and Follow the Perturbed Leader. We completely characterize the learning dynamics of mean-based algorithms, under two notions of convergence: (1) time-average: the fraction of rounds where bidders play a Nash equilibrium converges to 1; (2) last-iterate: the mixed strategy profile of bidders converges to a Nash equilibrium. Specifically, the results depend on the number of bidders with the highest value: - If the number is at least three, the dynamics almost surely converges to a Nash equilibrium of the auction, in both time-average and last-iterate. - If the number is two, the dynamics almost surely converges to a Nash equilibrium in time-average but not necessarily last-iterate. - If the number is one, the dynamics may not converge to a Nash equilibrium in time-average or last-iterate. Our discovery opens up new possibilities in the study of the convergence of learning dynamics.
format Preprint
id arxiv_https___arxiv_org_abs_2110_03906
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Nash Convergence of Mean-Based Learning Algorithms in First-Price Auctions
Deng, Xiaotie
Hu, Xinyan
Lin, Tao
Zheng, Weiqiang
Computer Science and Game Theory
Artificial Intelligence
Machine Learning
Multiagent Systems
Theoretical Economics
The convergence properties of learning dynamics in repeated auctions is a timely and important question, with numerous applications in, e.g., online advertising markets. This work focuses on repeated first-price auctions where bidders with fixed values learn to bid using mean-based algorithms -- a large class of online learning algorithms that include popular no-regret algorithms such as Multiplicative Weights Update and Follow the Perturbed Leader. We completely characterize the learning dynamics of mean-based algorithms, under two notions of convergence: (1) time-average: the fraction of rounds where bidders play a Nash equilibrium converges to 1; (2) last-iterate: the mixed strategy profile of bidders converges to a Nash equilibrium. Specifically, the results depend on the number of bidders with the highest value: - If the number is at least three, the dynamics almost surely converges to a Nash equilibrium of the auction, in both time-average and last-iterate. - If the number is two, the dynamics almost surely converges to a Nash equilibrium in time-average but not necessarily last-iterate. - If the number is one, the dynamics may not converge to a Nash equilibrium in time-average or last-iterate. Our discovery opens up new possibilities in the study of the convergence of learning dynamics.
title Nash Convergence of Mean-Based Learning Algorithms in First-Price Auctions
topic Computer Science and Game Theory
Artificial Intelligence
Machine Learning
Multiagent Systems
Theoretical Economics
url https://arxiv.org/abs/2110.03906