News Item

"A classical density functional from machine learning and a convolutional neural network" by Shang-Chun Lin and Martin Oettel has received the Select label. Warm congratulations to the authors!

This brilliant paper presents an inspiring conceptual DFT study, using machine learning methods to approximate a classical density functional, obtaining precise data for density distributions in given external potentials. In addition, they relate these simulations to their theoretical considerations, which encompass a variety of elements from classical liquid state theory to machine learning techniques. Conceptually, this goes far beyond what has been done in electronic structure DFT using machine learning. The hybrid approach of the current paper gives much insight into the formal (analytical) structure of the attractive contribution to the excess free energy functional. With the increasing use of machine learning, this study points towards the future direction on how to use it in the field of DFT.