19–23 May 2025
CERN
Europe/Zurich timezone

Enabling Scientific Discovery through Symbolic Models using Machine Learning

Not scheduled
20m
61/1-201 - Pas perdus - Not a meeting room - (CERN)

61/1-201 - Pas perdus - Not a meeting room -

CERN

10
Show room on map
Poster 9 ML in phenomenology and theory Poster Session

Speaker

Dr Nour Makke (Qatar Computing Research Institute)

Description

Machine learning is making its path into natural sciences. A key limitation in ML from a science perspective is the black-box nature of deep neural networks. An alternative is to learn succinct mathematical equations, thus interpretable models, directly from data, allowing for a deeper understanding and scientific reasoning, making the path toward new scientific discovery. Symbolic regression is the ML subfield to infer analytical models from data. It has gained a growing interest in recent years.

Most symbolic regression applications make use of synthetic data. However, recent applications used experimental data measured in high-energy physics and astronomy, showing promise in learning fundamental physics laws and phenomenological models in physics. We present a review of symbolic regression applications to high-energy physics data, highlight the main limitations of symbolic regression, and discuss new perspectives.

Artificial Intelligence Review 57 (1), 2
PNAS Nexus 3 (11) pgae467
Machine Learning: Science and Technology (accepted manuscript) DOI 10.1088/2632-2153/adb3ec, arXiv: 2501.07123
Oral talk at ACM KDD 2024, Barcelona
Oral talk at AAAI 2025, Philadelphia

Would you like to be considered for an oral presentation? Yes

Author

Dr Nour Makke (Qatar Computing Research Institute)

Presentation materials

There are no materials yet.