2 code implementations • 13 Feb 2024 • Rokas Elijošius, Fabian Zills, Ilyes Batatia, Sam Walton Norwood, Dávid Péter Kovács, Christian Holm, Gábor Csányi
Using insights from the trained model, we present Similarity-based Molecular Generation (SiMGen), a new method for zero shot molecular generation.
1 code implementation • 30 Jan 2024 • Ivan Grega, Ilyes Batatia, Gábor Csányi, Sri Karlapati, Vikram S. Deshpande
In this work, we generate a big dataset of structure-property relationships for strut-based lattices.
no code implementations • 16 Oct 2023 • Ilyes Batatia
To any coordinate-independent feature field on a manifold comes attached an equivariant embedding of the principal bundle to the space of numerical features.
no code implementations • 16 Oct 2023 • Ilyes Batatia, Lars L. Schaaf, Huajie Chen, Gábor Csányi, Christoph Ortner, Felix A. Faber
Graph Neural Networks (GNNs), especially message-passing neural networks (MPNNs), have emerged as powerful architectures for learning on graphs in diverse applications.
Ranked #3 on Graph Regression on ZINC-500k
no code implementations • 23 Jun 2023 • Jose M Munoz, Ilyes Batatia, Christoph Ortner, Francesco Romeo
Deep learning approaches for jet tagging in high-energy physics are characterized as black boxes that process a large amount of information from which it is difficult to extract key distinctive observables.
1 code implementation • 23 May 2023 • David Peter Kovacs, Ilyes Batatia, Eszter Sara Arany, Gabor Csanyi
We further demonstrate that the strictly local atom-centered model is sufficient for such tasks even in the case of large molecules and weakly interacting molecular assemblies.
no code implementations • 2 Oct 2022 • James P. Darby, Dávid P. Kovács, Ilyes Batatia, Miguel A. Caro, Gus L. W. Hart, Christoph Ortner, Gábor Csányi
Density based representations of atomic environments that are invariant under Euclidean symmetries have become a widely used tool in the machine learning of interatomic potentials, broader data-driven atomistic modelling and the visualisation and analysis of materials datasets. The standard mechanism used to incorporate chemical element information is to create separate densities for each element and form tensor products between them.
1 code implementation • 17 Jul 2022 • Jose M Munoz, Ilyes Batatia, Christoph Ortner
Deep Learning approaches are becoming the go-to methods for data analysis in High Energy Physics (HEP).
2 code implementations • 15 Jun 2022 • Ilyes Batatia, Dávid Péter Kovács, Gregor N. C. Simm, Christoph Ortner, Gábor Csányi
In particular, we show that using four-body messages reduces the required number of message passing iterations to just two, resulting in a fast and highly parallelizable model, reaching or exceeding state-of-the-art accuracy on the rMD17, 3BPA, and AcAc benchmark tasks.
2 code implementations • 13 May 2022 • Ilyes Batatia, Simon Batzner, Dávid Péter Kovács, Albert Musaelian, Gregor N. C. Simm, Ralf Drautz, Christoph Ortner, Boris Kozinsky, Gábor Csányi
The rapid progress of machine learning interatomic potentials over the past couple of years produced a number of new architectures.