AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Scaling Molecular Modeling with Deep Learning
This chapter discusses a new paper from DeepMind that explores the potential of using deep learning to predict new crystal structures. They outline the two-step process used in the project, involving generating various atomic structures and using a graph neural network to model the properties of these structures. The chapter concludes by highlighting the significance of this work in demonstrating the potential of scaling in molecular modeling.