Publications by authors named "M Kornbluth"
A simultaneously accurate and computationally efficient parametrization of the potential energy surface of molecules and materials is a long-standing goal in the natural sciences. While atom-centered message passing neural networks (MPNNs) have shown remarkable accuracy, their information propagation has limited the accessible length-scales. Local methods, conversely, scale to large simulations but have suffered from inferior accuracy.
View Article and Find Full Text PDFLeveraging ab initio data at scale has enabled the development of machine learning models capable of extremely accurate and fast molecular property prediction. A central paradigm of many previous studies focuses on generating predictions for only a fixed set of properties. Recent lines of research instead aim to explicitly learn the electronic structure via molecular wavefunctions, from which other quantum chemical properties can be directly derived.
View Article and Find Full Text PDFArticle Synopsis
- - NequIP is a new neural network model that uses E(3)-equivariant convolutions to learn interatomic potentials from quantum mechanical calculations, improving how atomic environments are represented.
- - This approach achieves top-tier accuracy across various molecules and materials, showing the ability to work effectively with much less training data than traditional models.
- - By requiring significantly fewer training samples, NequIP challenges the notion that deep learning networks need large datasets, enabling accurate simulations of molecular dynamics over extended time periods.
We present a novel multidisciplinary approach for the treatment of electrical storm combining bilateral cardiac sympathectomy, extrapericardial coil insertion, and implantable cardioverter defibrillator upgrade in a patient with nonischemic cardiomyopathy and ventricular arrhythmias refractory to conventional therapies. ().
View Article and Find Full Text PDFArticle Synopsis
- The study highlights the challenges of calculating ionic transport properties through traditional methods, which can be costly and inefficient.
- A new approach using spectral decomposition of ionic displacement covariance is introduced, offering a way to identify diffusion eigenmodes that capture the correlation structure of ionic movements.
- The method is validated with mathematical proofs and applied to practical electrolyte materials, showing its effectiveness in reducing uncertainty and speeding up calculations of ionic conductivity.