Parsimonious Learning-Augmented Caching
This work addresses the challenge of reducing prediction usage in learning-augmented caching algorithms, which is an incremental improvement for online algorithm design.
The paper tackles the problem of designing learning-augmented algorithms for caching by introducing a parsimonious approach that uses a sublinear number of predictions, achieving quantitatively similar results to prior methods.
Learning-augmented algorithms -- in which, traditional algorithms are augmented with machine-learned predictions -- have emerged as a framework to go beyond worst-case analysis. The overarching goal is to design algorithms that perform near-optimally when the predictions are accurate yet retain certain worst-case guarantees irrespective of the accuracy of the predictions. This framework has been successfully applied to online problems such as caching where the predictions can be used to alleviate uncertainties. In this paper we introduce and study the setting in which the learning-augmented algorithm can utilize the predictions parsimoniously. We consider the caching problem -- which has been extensively studied in the learning-augmented setting -- and show that one can achieve quantitatively similar results but only using a sublinear number of predictions.