Materialsvirtuallab.org Website Review

MATERIALS GRAPH NETWORKS AS A UNIVERSAL MACHINE LEARNING FRAMEWORK FOR MOLECULES AND CRYSTALS. approach.We show that MEGNet models are a universal approach to machine learning for both crystals and molecules, outperforming prior ML models on a broad array of properties. We also demonstrate the incorporation of state (e.g., temperature, pressure) into MEGNet models, and how transfer learning can be used to accelerate and improve the accuracy of models trained on smaller data sets. Our work (Wang, Z.; Ha, J.; Kim, Y. H.; Im, W. Bin; McKittrick, J.; Ong, S. P. Mining Unexplored Chemistries for Phosphors for Our former group member, Zhenbin Wang, has just been awarded the Gareth Thomas Materials Excellence Award 2018! This award As*ociate Director of the Institute of Mechanics and Materials at UC San Diego, and a leading Materials SCience of the 20th century. Congratulations to Zhenbin on this great honor and we wish him all the best in his postdoctoral stint at the group of Prof Jens Norskov in Weike’s paper on “De*p Neural Networks for Accurate Predictions of Crystal Stability” is now out in Nature Communications. Predicting the stability of crystals is one of the central problems in materials science. Here, we show that de*p neural networks, i.e., algorithms that mimic the an*mal brain, utilizing just the electronegativity and ionic radii as inputs can predict formation energies of crystals with extremely high accuracy. We also demonstrate how these models can be generalized for mixed crystals using a binary encoding scheme, and use News: UCSD News: Scientists use artificial neural networks to predict new stable materials Xiangguo’s article on “Quantum-accurate spectral neighbor an*lysis potential models for Ni-Mo binary alloys and fcc metals” has just been published in Physical Review B! In this work, we extend the spectral neighbor an*lysis potential, or SNAP, approach to fcc Ni-bcc Mo binary alloy systems. These new potentials are a substantial improvement over previous potentials based on the embedded atom method