LLMs learn governing principles of dynamical systems, revealing an in-context neural scaling law

1 code implementation • 1 Feb 2024 • Toni J. B. Liu, Nicolas Boullé, Raphaël Sarfati, Christopher J. Earls

Pretrained large language models (LLMs) are surprisingly effective at performing zero-shot tasks, including time-series forecasting.

Language Modelling Prompt Engineering +2

Bayesian Deep Learning for Partial Differential Equation Parameter Discovery with Sparse and Noisy Data

1 code implementation • 5 Aug 2021 • Christophe Bonneville, Christopher J. Earls

One caveat concerning current methods is the need for large amounts of ("clean") data, in order to characterize the full system response and discover underlying physical models.

Data-driven discovery of Green's functions with human-understandable deep learning

2 code implementations • 1 May 2021 • Nicolas Boullé, Christopher J. Earls, Alex Townsend

There is an opportunity for deep learning to revolutionize science and technology by revealing its findings in a human interpretable manner.

scientific discovery