Fast electron density estimation of molecules, liquids, and solids using neural networks

By Peter Bjørn Jørgensen, DTU Energy, Denmark
ICN2 Seminar: Fast electron density estimation of molecules, liquids, and solids using neural networks
Date: February 2023
Abstract: In this talk I will present DeepDFT, a machine learning framework for the prediction of the electron density. I will give a brief introduction to deep learning for physics, leading into recent work on graph neural networks on which the model is based upon. The model is tested on multiple datasets of molecules (QM9), liquid ethylene carbonate electrolyte and lithium ion battery cathode material (NMC). The accuracy on all three datasets is beyond state of the art and the computation time is orders of magnitude faster than DFT.
Hosted by Prof. Pablo Ordejón,Theory and Simulation Group Leader at ICN2.