What is it about?

Rogue waves are rare ocean waves that routinely damage ships and offshore structures. We use artificial intelligence to transform a vast database of wave observations into a concise mathematical equation for the occurrence probability of rogue waves. This equation can be analyzed and incorporated into the research canon by human scientists, and paves the way for a better rogue wave forecast.

Featured Image

Why is it important?

Machine learning models excel at pattern recognition and are used extensively in predictive science and engineering. But because of their black-box nature, better machine learning models do not always lead to greater human understanding. This study demonstrates the potential of using AI as an assistant for semi-automated scientific discovery from data in a real-world application.


I believe that scientific machine learning should first and foremost be aimed at increasing human understanding. I am thrilled that we were able to show how AI can augment the scientific method instead of using it to create models that we can't understand. I hope that other scientists will recognize the potential to use machine learning as a tool to convert large amounts of data into insight.

Dion Häfner
Pasteur Labs

Read the Original

This page is a summary of: Machine-guided discovery of a real-world rogue wave model, Proceedings of the National Academy of Sciences, November 2023, Proceedings of the National Academy of Sciences,
DOI: 10.1073/pnas.2306275120.
You can read the full text:



The following have contributed to this page