Machine Learning for Science: Mathematics at the Interface of Data-driven and Mechanistic Modelling

Jessica Montgomery; Neil D. Lawrence

doi:10.4171/OWR/2023/26

Back to publications

Machine Learning for Science: Mathematics at the Interface of Data-driven and Mechanistic Modelling

Jessica Montgomery, Neil D. Lawrence

Oberwolfach Reports, EMS Press 20(2), 2026.

Abstract

Rapid progress in machine learning is enabling scientific advances across a range of disciplines. However, the utility of machine learning for science remains constrained by its current inability to translate insights from data about the dynamics of a system to new scientific knowledge about why those dynamics emerge, as traditionally represented by physical modelling. Mathematics is the interface that bridges data-driven and physical models of the world and can provide a foundation for delivering such knowledge. This workshop convened researchers working across domains with a shared interest in mathematics, machine learning, and their application in the sciences, to explore how tools of mathematics can help build machine learning tools for scientific discovery.

Links

Cite this Paper

BibTeX


@Article{publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling,
  title = 	 {Machine Learning for Science: Mathematics at the Interface of Data-driven and Mechanistic Modelling},
  author = 	 {Montgomery, Jessica and Lawrence, Neil D.},
  journal =      {Oberwolfach Reports},
  year = 	 {2026},
  volume = 	 {20},
  number =       {2},
  publisher =    {EMS Press},
  doi = 	 {10.4171/OWR/2023/26},
  url = 	 {/publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling.html},
  abstract = 	 {Rapid progress in machine learning is enabling scientific advances across a range of disciplines. However, the utility of machine learning for science remains constrained by its current inability to translate insights from data about the dynamics of a system to new scientific knowledge about why those dynamics emerge, as traditionally represented by physical modelling. Mathematics is the interface that bridges data-driven and physical models of the world and can provide a foundation for delivering such knowledge. This workshop convened researchers working across domains with a shared interest in mathematics, machine learning, and their application in the sciences, to explore how tools of mathematics can help build machine learning tools for scientific discovery.}
}

Endnote

%0 Journal Article
%T Machine Learning for Science: Mathematics at the Interface of Data-driven and Mechanistic Modelling
%A Jessica Montgomery
%A Neil D. Lawrence
%J Oberwolfach Reports
%D 2026	
%F publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling
%I EMS Press
%R 10.4171/OWR/2023/26
%U /publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling.html
%V 20
%N 2
%X Rapid progress in machine learning is enabling scientific advances across a range of disciplines. However, the utility of machine learning for science remains constrained by its current inability to translate insights from data about the dynamics of a system to new scientific knowledge about why those dynamics emerge, as traditionally represented by physical modelling. Mathematics is the interface that bridges data-driven and physical models of the world and can provide a foundation for delivering such knowledge. This workshop convened researchers working across domains with a shared interest in mathematics, machine learning, and their application in the sciences, to explore how tools of mathematics can help build machine learning tools for scientific discovery.

RIS


TY  - JOUR
TI  - Machine Learning for Science: Mathematics at the Interface of Data-driven and Mechanistic Modelling
AU  - Jessica Montgomery
AU  - Neil D. Lawrence
DA  - 2026/09/29	
ID  - publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling
PB  - EMS Press
VL  - 20
IS  - 2
DO  - 10.4171/OWR/2023/26
UR  - /publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling.html
AB  - Rapid progress in machine learning is enabling scientific advances across a range of disciplines. However, the utility of machine learning for science remains constrained by its current inability to translate insights from data about the dynamics of a system to new scientific knowledge about why those dynamics emerge, as traditionally represented by physical modelling. Mathematics is the interface that bridges data-driven and physical models of the world and can provide a foundation for delivering such knowledge. This workshop convened researchers working across domains with a shared interest in mathematics, machine learning, and their application in the sciences, to explore how tools of mathematics can help build machine learning tools for scientific discovery.
ER  -

APA


Montgomery, J. & Lawrence, N.D.. (2026). Machine Learning for Science: Mathematics at the Interface of Data-driven and Mechanistic Modelling. Oberwolfach Reports 20(2) doi:10.4171/OWR/2023/26 Available from /publications/machine-learning-for-science-mathematics-at-the-interface-of-data-driven-and-mechanistic-modelling.html.

Related People

Jess Montgomery

Director, ai@cam and Accelerate Science, Cambridge University

View Profile

Neil D. Lawrence

The DeepMind Professor of Machine Learning, Cambridge University

View Profile