Learning Concave Bid Shading Strategies in Online Auctions via Measure-valued Proximal Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nodozi, Iman, Gligorijevic, Djordje, Halder, Abhishek
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912585442394112
author Nodozi, Iman
Gligorijevic, Djordje
Halder, Abhishek
author_facet Nodozi, Iman
Gligorijevic, Djordje
Halder, Abhishek
contents This work proposes a bid shading strategy for first-price auctions as a measure-valued optimization problem. We consider a standard parametric form for bid shading and formulate the problem as convex optimization over the joint distribution of shading parameters. After each auction, the shading parameter distribution is adapted via a regularized Wasserstein-proximal update with a data-driven energy functional. This energy functional is conditional on the context, i.e., on publisher/user attributes such as domain, ad slot type, device, or location. The proposed algorithm encourages the bid distribution to place more weight on values with higher expected surplus, i.e., where the win probability and the value gap are both large. We show that the resulting measure-valued convex optimization problem admits a closed form solution. A numerical example illustrates the proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2509_10693
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Concave Bid Shading Strategies in Online Auctions via Measure-valued Proximal Optimization
Nodozi, Iman
Gligorijevic, Djordje
Halder, Abhishek
Machine Learning
Artificial Intelligence
Optimization and Control
This work proposes a bid shading strategy for first-price auctions as a measure-valued optimization problem. We consider a standard parametric form for bid shading and formulate the problem as convex optimization over the joint distribution of shading parameters. After each auction, the shading parameter distribution is adapted via a regularized Wasserstein-proximal update with a data-driven energy functional. This energy functional is conditional on the context, i.e., on publisher/user attributes such as domain, ad slot type, device, or location. The proposed algorithm encourages the bid distribution to place more weight on values with higher expected surplus, i.e., where the win probability and the value gap are both large. We show that the resulting measure-valued convex optimization problem admits a closed form solution. A numerical example illustrates the proposed method.
title Learning Concave Bid Shading Strategies in Online Auctions via Measure-valued Proximal Optimization
topic Machine Learning
Artificial Intelligence
Optimization and Control
url https://arxiv.org/abs/2509.10693