Bridging Machine Learning and Sciences: Opportunities and Challenges

TL;DR

This paper reviews how machine learning, especially anomaly detection and out-of-distribution methods, is increasingly applied in scientific fields, highlighting opportunities, challenges, and the need for interdisciplinary research.

Contribution

It critically examines the application of ML techniques in sciences, emphasizing transferable practices and domain-specific challenges for future interdisciplinary research.

Findings

ML techniques show potential in scientific data analysis

Challenges include data universality and model robustness

Transferable practices can bridge ML and sciences

Abstract

The application of machine learning in sciences has seen exciting advances in recent years. As a widely applicable technique, anomaly detection has been long studied in the machine learning community. Especially, deep neural nets-based out-of-distribution detection has made great progress for high-dimensional data. Recently, these techniques have been showing their potential in scientific disciplines. We take a critical look at their applicative prospects including data universality, experimental protocols, model robustness, etc. We discuss examples that display transferable practices and domain-specific challenges simultaneously, providing a starting point for establishing a novel interdisciplinary research paradigm in the near future.