The Pareto Frontier of Randomized Learning-Augmented Online Bidding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Degryse, Mathis, Saakour, Imrane, Dürr, Christoph, Angelopoulos, Spyros
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918500569710592
author Degryse, Mathis
Saakour, Imrane
Dürr, Christoph
Angelopoulos, Spyros
author_facet Degryse, Mathis
Saakour, Imrane
Dürr, Christoph
Angelopoulos, Spyros
contents Online bidding is a classical problem in online decision-making, with applications in resource allocation, hierarchical clustering, and the analysis of approximation algorithms. We study its randomized learning-augmented variant, where an online algorithm generates a sequence of random bids while leveraging predictions from an oracle. We provide analytical upper and lower bounds on the optimal consistency $C$ as a function of the robustness $R$, which match when $R \geq 2.885$, effectively closing the gap left by previous work. The key technical ingredient is the notion of a bidding function, a novel abstraction that provides a unified framework for the design and analysis of randomized bidding strategies. We complement our theoretical results with an experimental application of randomized bidding to the incremental median problem, demonstrating the applicability of our algorithm in practical clustering settings.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06106
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Pareto Frontier of Randomized Learning-Augmented Online Bidding
Degryse, Mathis
Saakour, Imrane
Dürr, Christoph
Angelopoulos, Spyros
Data Structures and Algorithms
F.2.0
Online bidding is a classical problem in online decision-making, with applications in resource allocation, hierarchical clustering, and the analysis of approximation algorithms. We study its randomized learning-augmented variant, where an online algorithm generates a sequence of random bids while leveraging predictions from an oracle. We provide analytical upper and lower bounds on the optimal consistency $C$ as a function of the robustness $R$, which match when $R \geq 2.885$, effectively closing the gap left by previous work. The key technical ingredient is the notion of a bidding function, a novel abstraction that provides a unified framework for the design and analysis of randomized bidding strategies. We complement our theoretical results with an experimental application of randomized bidding to the incremental median problem, demonstrating the applicability of our algorithm in practical clustering settings.
title The Pareto Frontier of Randomized Learning-Augmented Online Bidding
topic Data Structures and Algorithms
F.2.0
url https://arxiv.org/abs/2605.06106