Auctions Meet Bandits: An Empirical Analysis

2025-08-28

AI Summary

Emerging advertisers in sponsored search advertising face a quality score cold-start problem, compounded by the absence of principled exploration strategies in auction mechanisms. Method: We propose a unified auction-bandit framework that integrates Thompson Sampling into second-price auctions to dynamically learn quality scores, and design a keyword-feature-driven adaptive exploration intensity control policy to construct a revenue–efficiency Pareto frontier. Contribution/Results: Theoretically, we characterize fundamental distinctions between auction-bandit coupled systems and classical bandit problems. Empirically, evaluations on real-world mobile app store keyword advertising data demonstrate significant improvements in platform revenue and ad allocation efficiency—achieving quantifiable, substantial gains on both metrics. Our approach yields actionable policy insights for mechanism design in practical sponsored search settings.

Technology Category

• Machine Learning: Online Learning & Bandits
• Search and Optimization: Learning to Search
• Game Theory and Economic Paradigms: Auctions and Market-Based Systems

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Advertising auctions, pricing, markets, and exchanges
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

Sponsored search positions are typically allocated through real-time auctions, where the outcomes depend on advertisers' quality-adjusted bids - the product of their bids and quality scores. Although quality scoring helps promote ads with higher conversion outcomes, setting these scores for new advertisers in any given market is challenging, leading to the cold-start problem. To address this, platforms incorporate multi-armed bandit algorithms in auctions to balance exploration and exploitation. However, little is known about the optimal exploration strategies in such auction environments. We utilize data from a leading Asian mobile app store that places sponsored ads for keywords. The platform employs a Thompson Sampling algorithm within a second-price auction to learn quality scores and allocate a single sponsored position for each keyword. We empirically quantify the gains from optimizing exploration under this combined auction-bandit model and show that this problem differs substantially from the canonical bandit problem. Drawing on these empirical insights, we propose a customized exploration strategy in which the platform adjusts the exploration levels for each keyword according to its characteristics. We derive the Pareto frontier for revenue and efficiency and provide actionable policies, demonstrating substantial gains for the platform on both metrics when using a tailored exploration approach.

Problem

• Addressing cold-start problem for new advertisers via auction-bandit integration
• Optimizing exploration strategies in real-time sponsored search auctions
• Balancing revenue and efficiency through customized keyword exploration levels

Innovation

• Thompson Sampling in second-price auctions
• Customized exploration strategy per keyword
• Pareto frontier optimization for revenue efficiency