SC-Square: Future Progress with Machine Learning?

TL;DR

This paper surveys recent efforts to leverage Machine Learning techniques to enhance algorithms relevant to the SC-Square community, focusing on efficiency and implementation choices.

Contribution

It provides an overview of how Machine Learning is being used to improve algorithm efficiency and tractability in high-performance computing.

Findings

Machine Learning can optimize algorithm implementation choices.

ML approaches impact efficiency and tractability of algorithms.

Survey highlights recent advances and future directions.

Abstract

The algorithms employed by our communities are often underspecified, and thus have multiple implementation choices, which do not effect the correctness of the output, but do impact the efficiency or even tractability of its production. In this extended abstract, to accompany a keynote talk at the 2021 SC-Square Workshop, we survey recent work (both the author's and from the literature) on the use of Machine Learning technology to improve algorithms of interest to SC-Square.