Séminaire/Seminar – Simulating matter on the quantum scale with Artificial Intelligence – David Turban – 28/09/22

13 h 30 min
INSP - Sorbonne Université
Sorbonne Université Campus Pierre et Marie Curie 4 place Jussieu, Paris, 75005
Type d’évènement/Event category

Mercredi 28 septembre 2022 à 13 h 30.

INSP – Salle 22-23 317


David Turban – chercheur chez Google Deepmind depuis 2018


Density functional theory describes matter at the quantum level, but all popular approximations suffer from systematic errors that arise from the violation of mathematical properties of the exact functional. We overcame this fundamental limitation by training a neural network on molecular data and on fictitious systems with fractional charge and spin. The resulting functional, DM21 (DeepMind 21), correctly describes typical examples of artificial charge delocalization and strong correlation and performs better than traditional functionals on thorough benchmarks for main-group atoms and molecules. DM21 accurately models complex systems such as hydrogen chains, charged DNA base pairs, and diradical transition states. More crucially for the field, because our methodology relies on data and constraints, which are continually improving, it represents a viable pathway toward the exact universal functional.