Addressing Delayed Feedback in Conversion Rate Prediction via Influence Functions

2025-02-01

AI Summary

In online advertising, conversion rate (CVR) prediction is severely hindered by delayed feedback—ranging from hours to weeks—which induces incomplete training data, model bias, and degraded performance. To address this, we propose the first application of influence functions to delay-aware CVR modeling, enabling efficient, retraining-free online parameter adaptation. Specifically, we reformulate the inverse-Hessian-vector product—a core component of influence estimation—as a scalable optimization problem, drastically reducing computational overhead. Evaluated on multiple benchmark datasets, our method consistently outperforms state-of-the-art approaches, achieving higher CVR prediction accuracy and improved adaptability to long-term user behavior evolution. These advances yield more robust decision support for cost-per-action (CPA) bidding systems.

Technology Category

• Machine Learning: Online Learning & Bandits
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In the realm of online digital advertising, conversion rate (CVR) prediction plays a pivotal role in maximizing revenue under cost-per-conversion (CPA) models, where advertisers are charged only when users complete specific actions, such as making a purchase. A major challenge in CVR prediction lies in the delayed feedback problem-conversions may occur hours or even weeks after initial user interactions. This delay complicates model training, as recent data may be incomplete, leading to biases and diminished performance. Although existing methods attempt to address this issue, they often fall short in adapting to evolving user behaviors and depend on auxiliary models, which introduces computational inefficiencies and the risk of model inconsistency. In this work, we propose an Influence Function-empowered framework for Delayed Feedback Modeling (IF-DFM). IF-DFM leverages influence functions to estimate how newly acquired and delayed conversion data impact model parameters, enabling efficient parameter updates without the need for full retraining. Additionally, we present a scalable algorithm that efficiently computes parameter updates by reframing the inverse Hessian-vector product as an optimization problem, striking a balance between computational efficiency and effectiveness. Extensive experiments on benchmark datasets demonstrate that IF-DFM consistently surpasses state-of-the-art methods, significantly enhancing both prediction accuracy and model adaptability.

Problem

• Address delayed feedback in CVR prediction
• Improve model training with incomplete data
• Enhance computational efficiency and accuracy

Innovation

• Influence functions for delayed feedback
• Scalable parameter update algorithm
• Optimization-based inverse Hessian-vector product