Improving Online Rent-or-Buy Algorithms with Sequential Decision Making and ML Predictions
Abstract
In this work we study online rent-or-buy problems as a sequential decision making problem. We show how one can integrate predictions, typically coming from a machine learning (ML) setup, into this framework. Specifically, we consider the ski-rental problem and the dynamic TCP acknowledgment problem. We present new online algorithms and obtain explicit performance bounds in-terms of the accuracy of the prediction. Our algorithms are close to optimal with accurate predictions while hedging against less accurate predictions.
Cite
Text
Banerjee. "Improving Online Rent-or-Buy Algorithms with Sequential Decision Making and ML Predictions." Neural Information Processing Systems, 2020.Markdown
[Banerjee. "Improving Online Rent-or-Buy Algorithms with Sequential Decision Making and ML Predictions." Neural Information Processing Systems, 2020.](https://mlanthology.org/neurips/2020/banerjee2020neurips-improving/)BibTeX
@inproceedings{banerjee2020neurips-improving,
title = {{Improving Online Rent-or-Buy Algorithms with Sequential Decision Making and ML Predictions}},
author = {Banerjee, Shom},
booktitle = {Neural Information Processing Systems},
year = {2020},
url = {https://mlanthology.org/neurips/2020/banerjee2020neurips-improving/}
}