Customizing ML predictions for online algorithms

Conference Paper

A popular line of recent research incorporates ML advice in the design of online algorithms to improve their performance in typical instances. These papers treat the ML algorithm as a blackbox, and redesign online algorithms to take advantage of ML predictions. In this paper, we ask the complementary question: can we redesign ML algorithms to provide better predictions for online algorithms? We explore this question in the context of the classic rent-or-buy problem, and show that incorporating optimization benchmarks in ML loss functions leads to signifcantly better performance, while maintaining a worst-case adversarial result when the advice is completely wrong. We support this fnding both through theoretical bounds and numerical simulations.

Duke Authors

Cited Authors

  • Anand, K; Ge, R

Published Date

  • January 1, 2020

Published In

  • 37th International Conference on Machine Learning, Icml 2020

Volume / Issue

  • PartF168147-1 /

Start / End Page

  • 280 - 290

International Standard Book Number 13 (ISBN-13)

  • 9781713821120

Citation Source

  • Scopus