Competitive caching with machine learned advice

TL;DR

This paper introduces a framework that combines online algorithms with machine learned advice to improve caching strategies, achieving better competitive ratios when the oracle's predictions are accurate, and demonstrating strong empirical performance.

Contribution

The work develops a black-box method to augment online algorithms with machine learning advice, improving competitive ratios in caching problems based on prediction accuracy.

Findings

The modified Marker algorithm's competitive ratio improves as oracle error decreases.

The combined approach always maintains an $O(\log k)$ competitive ratio without oracle input.

Empirical results show strong performance with simple machine learning predictions.

Abstract

Traditional online algorithms encapsulate decision making under uncertainty, and give ways to hedge against all possible future events, while guaranteeing a nearly optimal solution as compared to an offline optimum. On the other hand, machine learning algorithms are in the business of extrapolating patterns found in the data to predict the future, and usually come with strong guarantees on the expected generalization error. In this work we develop a framework for augmenting online algorithms with a machine learned oracle to achieve competitive ratios that provably improve upon unconditional worst case lower bounds when the oracle has low error. Our approach treats the oracle as a complete black box, and is not dependent on its inner workings, or the exact distribution of its errors. We apply this framework to the traditional caching problem -- creating an eviction strategy for a cache of size $k$. We demonstrate that naively following the oracle's recommendations may lead to very poor performance, even when the average error is quite low. Instead we show how to modify the Marker algorithm to take into account the oracle's predictions, and prove that this combined approach achieves a competitive ratio that both (i) decreases as the oracle's error decreases, and (ii) is always capped by $O(\log k)$, which can be achieved without any oracle input. We complement our results with an empirical evaluation of our algorithm on real world datasets, and show that it performs well empirically even using simple off-the-shelf predictions.