English

Near-Optimal Consistency-Robustness Trade-Offs for Learning-Augmented Online Knapsack Problems

Machine Learning 2025-07-10 v2 Computer Science and Game Theory

Abstract

This paper introduces a family of learning-augmented algorithms for online knapsack problems that achieve near Pareto-optimal consistency-robustness trade-offs through a simple combination of trusted learning-augmented and worst-case algorithms. Our approach relies on succinct, practical predictions -- single values or intervals estimating the minimum value of any item in an offline solution. Additionally, we propose a novel fractional-to-integral conversion procedure, offering new insights for online algorithm design.

Keywords

Cite

@article{arxiv.2406.18752,
  title  = {Near-Optimal Consistency-Robustness Trade-Offs for Learning-Augmented Online Knapsack Problems},
  author = {Mohammadreza Daneshvaramoli and Helia Karisani and Adam Lechowicz and Bo Sun and Cameron Musco and Mohammad Hajiesmaili},
  journal= {arXiv preprint arXiv:2406.18752},
  year   = {2025}
}

Comments

31 pages, 16 figures, Accepted at ICML 2025

R2 v1 2026-06-28T17:20:34.914Z