Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

CPP: Fachverband Chemische Physik und Polymerphysik

CPP 33: Modeling and Simulation of Soft Matter IV

CPP 33.4: Vortrag

Donnerstag, 20. März 2025, 12:30–12:45, H34

Protonated water clusters by stochastic approaches: probing machine learning resilience against quantum Monte Carlo noise — •Matteo Peria, Antonino Marco Saitta, and Michele Casula — Sorbonne Université, 4 place Jussieu Paris, France

A complete understanding of the hydrogen bond and proton transfer mechanism in water is still lacking, since it requires an accurate potential energy surface (PES) and very expensive quantum mechanical simulations of the nuclear part. Reproducing this high-dimensional surface with current high-level computational chemistry methods is infeasible for the largest clusters. We test gradient-based kernel ridge regression methods and neural networks to reproduce the PES starting from a dataset of energies and forces of the protonated water clusters obtained via simulations combining classical molecular dynamics (MD) for the nuclei and quantum Monte Carlo (QMC) for the electrons. The QMC+MD approach yields very accurate results for the classical dynamics, which are however affected by the intrinsic noise inherent in the stochastic sampling of both nuclear and electronic phase space. We prove that QMC multivariate noise is not necessarily detrimental to the learning of energies and forces and we determine under which conditions one can derive accurate and reliable MLIPs from QMC data.

Keywords: quantum Monte Carlo; machine learning; protonated water; Grotthuss mechanism