Uncertainty Quantification of Click and Conversion Estimates for the Autobidding

2026-03-02

AI Summary

This work addresses the degradation of auction efficiency in automated bidding systems caused by noise in click-through rate (CTR) and conversion rate (CVR) predictions. To mitigate this issue, the authors propose DenoiseBid, a novel approach that introduces a Bayesian denoising mechanism into automated bidding for the first time. By leveraging Bayesian inference and distribution recovery techniques, DenoiseBid reconstructs the true posterior distribution from noisy predictions and integrates uncertainty estimates from pre-trained models to calibrate bidding signals. Extensive experiments on synthetic data as well as real-world datasets from iPinYou and BAT demonstrate the effectiveness of the method: DenoiseBid significantly improves bidding efficiency and exhibits strong robustness across varying levels of prediction noise.

Technology Category

• Machine Learning: Online Learning & Bandits
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

Modern e-commerce platforms employ various auction mechanisms to allocate paid slots for a given item. To scale this approach to the millions of auctions, the platforms suggest promotion tools based on the autobidding algorithms. These algorithms typically depend on the Click-Through-Rate (CTR) and Conversion-Rate (CVR) estimates provided by a pre-trained machine learning model. However, the predictions of such models are uncertain and can significantly affect the performance of the autobidding algorithm. To address this issue, we propose the DenoiseBid method, which corrects the generated CTRs and CVRs to make the resulting bids more efficient in auctions. The underlying idea of our method is to employ a Bayesian approach and replace noisy CTR or CVR estimates with those from recovered distributions. To demonstrate the performance of the proposed approach, we perform extensive experiments on the synthetic, iPinYou, and BAT datasets. To evaluate the robustness of our approach to the noise scale, we use synthetic noise and noise estimated from the predictions of the pre-trained machine learning model.

Problem

• Uncertainty Quantification
• Click-Through-Rate
• Conversion-Rate
• Autobidding
• Auction Mechanism

Innovation

• Uncertainty Quantification
• Autobidding
• Bayesian Denoising
• CTR/CVR Estimation
• Robust Bidding