Neural-network based simulation of rare event processes at the water/oxide interface

Subproject P12

Atomistic computer simulations of processes occurring at the water/oxide interface are challenging in several ways. The calculation of atomic forces based on ab initio methods is computationally very demanding, and barrier crossing events may lead to long computation times. Both these aspects severely limit accessible system sizes and simulation times.

In project P12, we will address these challenges using a combination of machine learning and advanced rare event sampling methods. In particular, using software developed in our group and collaborating with P03 Kresse, we will train neural network potentials based on the Behler-Parrinello approach for oxide/water interfaces, starting with the Fe₃O₄/water system studied in P11 Backus. We will pay special attention to error estimation and the correct treatment of long-range interactions. With the new potential, we will study the structure and dynamics of water near the oxide surface to provide the atomistic information necessary to rationalize the spectroscopy experiments of P11 Backus. Another important goal of P12 is to explore how deep generative models can be used to enhance rare event simulations. For this purpose, we will apply normalizing flows, represented by deep neural networks, to trajectory space. The resulting improved transition path sampling simulations will be used to study reactive processes investigated experimentally in other subprojects of TACO.

Christoph Dellago
PI

Expertise

Our research efforts focus on the development of simulation algorithms and their application to investigate dynamical processes in condensed matter systems based on the principles of equilibrium and non-equilibrium statistical mechanics. In particular, we have helped to create the transition path sampling methodology for the simulation of rare but important events, such as nucleation aprocesses, chemical reactions and biomolecular reorganizations. More recently, we have worked on applying machine learning methods to molecular structure recognition and the representation of potential and free energy surfaces.

Recent research topics include:

Self-assembly of nanocrystals
Folding and unfolding of biopolymers
Interfaces in aqueous systems
Phase separation in alloys
Thermo-polarisation
Structure and dynamics of water and ice
Cavitation
Crystallization
Non-equilibrium work fluctuations

Team

Christoph Dellago
PI

Pablo Montero de Hijes
PostDoc

Salvatore Romano
PhD Student

Associates

Alessandro Coretti
PostDoc

Andreas Tröster
PostDoc

Sebastian Falkner
PhD Student

Publications

2023

Jung, Hendrik; Covino, Roberto; Arjun, A.; Leitold, Christian; Dellago, Christoph; Bolhuis, Peter G.; Hummer, Gerhard

Machine-guided path sampling to discover mechanisms of molecular self-organization

Journal ArticleOpen Access

In: Nature Computational Science, vol. 3, no. 4, pp. 334–345, 2023.

Abstract | Links | BibTeX | Tags: P12

2021

Bircher, Martin P; Singraber, Andreas; Dellago, Christoph

Improved description of atomic environments using low-cost polynomial functions with compact support

Journal ArticleOpen Access

In: Machine Learning: Science and Technology, vol. 2, no. 3, pp. 035026, 2021.

Abstract | Links | BibTeX | Tags: P12, pre-TACO

@article{Bircher2021,

title = {Improved description of atomic environments using low-cost polynomial functions with compact support},

author = {Martin P Bircher and Andreas Singraber and Christoph Dellago},

doi = {10.1088/2632-2153/abf817},

year  = {2021},

date = {2021-06-16},

urldate = {2021-06-16},

journal = {Machine Learning: Science and Technology},

volume = {2},

number = {3},

pages = {035026},

publisher = {IOP Publishing},

abstract = {The prediction of chemical properties using machine learning techniques calls for a set of appropriate descriptors that accurately describe atomic and, on a larger scale, molecular environments. A mapping of conformational information on a space spanned by atom-centred symmetry functions (SF) has become a standard technique for energy and force predictions using high-dimensional neural network potentials (HDNNP). An appropriate choice of SFs is particularly crucial for accurate force predictions. Established atom-centred SFs, however, are limited in their flexibility, since their functional form restricts the angular domain that can be sampled without introducing problematic derivative discontinuities. Here, we introduce a class of atom-centred SFs based on polynomials with compact support called polynomial symmetry functions (PSF), which enable a free choice of both, the angular and the radial domain covered. We demonstrate that the accuracy of PSFs is either on par or considerably better than that of conventional, atom-centred SFs. In particular, a generic set of PSFs with an intuitive choice of the angular domain inspired by organic chemistry considerably improves prediction accuracy for organic molecules in the gaseous and liquid phase, with reductions in force prediction errors over a test set approaching 50% for certain systems. Contrary to established atom-centred SFs, computation of PSF does not involve any exponentials, and their intrinsic compact support supersedes use of separate cutoff functions, facilitating the choice of their free parameters. Most importantly, the number of floating point operations required to compute polynomial SFs introduced here is considerably lower than that of other state-of-the-art SFs, enabling their efficient implementation without the need of highly optimised code structures or caching, with speedups with respect to other state-of-the-art SFs reaching a factor of 4.5 to 5. This low-effort performance benefit substantially simplifies their use in new programs and emerging platforms such as graphical processing units. Overall, polynomial SFs with compact support improve accuracy of both, energy and force predictions with HDNNPs while enabling significant speedups compared to their well-established counterparts.},

keywords = {P12, pre-TACO},

pubstate = {published},

tppubtype = {article}

}

The prediction of chemical properties using machine learning techniques calls for a set of appropriate descriptors that accurately describe atomic and, on a larger scale, molecular environments. A mapping of conformational information on a space spanned by atom-centred symmetry functions (SF) has become a standard technique for energy and force predictions using high-dimensional neural network potentials (HDNNP). An appropriate choice of SFs is particularly crucial for accurate force predictions. Established atom-centred SFs, however, are limited in their flexibility, since their functional form restricts the angular domain that can be sampled without introducing problematic derivative discontinuities. Here, we introduce a class of atom-centred SFs based on polynomials with compact support called polynomial symmetry functions (PSF), which enable a free choice of both, the angular and the radial domain covered. We demonstrate that the accuracy of PSFs is either on par or considerably better than that of conventional, atom-centred SFs. In particular, a generic set of PSFs with an intuitive choice of the angular domain inspired by organic chemistry considerably improves prediction accuracy for organic molecules in the gaseous and liquid phase, with reductions in force prediction errors over a test set approaching 50% for certain systems. Contrary to established atom-centred SFs, computation of PSF does not involve any exponentials, and their intrinsic compact support supersedes use of separate cutoff functions, facilitating the choice of their free parameters. Most importantly, the number of floating point operations required to compute polynomial SFs introduced here is considerably lower than that of other state-of-the-art SFs, enabling their efficient implementation without the need of highly optimised code structures or caching, with speedups with respect to other state-of-the-art SFs reaching a factor of 4.5 to 5. This low-effort performance benefit substantially simplifies their use in new programs and emerging platforms such as graphical processing units. Overall, polynomial SFs with compact support improve accuracy of both, energy and force predictions with HDNNPs while enabling significant speedups compared to their well-established counterparts.