Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-15849
For citation please use:
Main Title: High-fidelity molecular dynamics trajectory reconstruction with bi-directional neural networks
Author(s): Winkler, Ludwig
Müller, Klaus-Robert
Sauceda, Huziel E.
Type: Article
URI: https://depositonce.tu-berlin.de/handle/11303/17070
http://dx.doi.org/10.14279/depositonce-15849
License: https://creativecommons.org/licenses/by/4.0/
Abstract: Molecular dynamics (MD) simulations are a cornerstone in science, enabling the investigation of a system’s thermodynamics all the way to analyzing intricate molecular interactions. In general, creating extended molecular trajectories can be a computationally expensive process, for example, when running ab-initio simulations. Hence, repeating such calculations to either obtain more accurate thermodynamics or to get a higher resolution in the dynamics generated by a fine-grained quantum interaction can be time- and computational resource-consuming. In this work, we explore different machine learning methodologies to increase the resolution of MD trajectories on-demand within a post-processing step. As a proof of concept, we analyse the performance of bi-directional neural networks (NNs) such as neural ODEs, Hamiltonian networks, recurrent NNs and long short-term memories, as well as the uni-directional variants as a reference, for MD simulations (here: the MD17 dataset). We have found that Bi-LSTMs are the best performing models; by utilizing the local time-symmetry of thermostated trajectories they can even learn long-range correlations and display high robustness to noisy dynamics across molecular complexity. Our models can reach accuracies of up to 10−4 Å in trajectory interpolation, which leads to the faithful reconstruction of several unseen high-frequency molecular vibration cycles. This renders the comparison between the learned and reference trajectories indistinguishable. The results reported in this work can serve (1) as a baseline for larger systems, as well as (2) for the construction of better MD integrators.
Subject(s): super-resolution
molecular dynamics
bi-directional recurrent neural networks
trajectory learning
LSTM
machine learning
molecular systems
Issue Date: 30-May-2022
Date Available: 9-Jun-2022
Language Code: en
DDC Class: 004 Datenverarbeitung; Informatik
Sponsor/Funder: BMBF, 01IS14013A, Verbundprojekt: BBDC - Berliner Kompetenzzentrum für Big Data
BMBF, 01GQ1115, D-JPN Verbund: Adaptive Gehirn-Computer-Schnittstellen (BCI) in nichtstationären Umgebungen
BMBF, 01GQ0850, Verbundprojekt: Bernstein Fokus Neurotechnologie - Nichtinvasive Neurotechnologie für Mensch-Maschine Interaktion - Teilprojekte A1, A3, A4, B4, W3, Zentrum
BMBF, 01IS18025A, Verbundprojekt BIFOLD-BBDC: Berlin Institute for the Foundations of Learning and Data
BMBF, 01IS18037A, Verbundprojekt BIFOLD-BZML: Berlin Institute for the Foundations of Learning and Data
DFG, 390685689, EXC 2046: MATH+: Berlin Mathematics Research Center
Journal Title: Machine Learning: Science and Technology
Publisher: IOP
Volume: 3
Issue: 2
Article Number: 025011
Publisher DOI: 10.1088/2632-2153/ac6ec6
EISSN: 2632-2153
TU Affiliation(s): Fak. 4 Elektrotechnik und Informatik » Inst. Softwaretechnik und Theoretische Informatik » FG Maschinelles Lernen
Appears in Collections:Technische Universität Berlin » Publications

Files in This Item:
mlst_3_2_025011.pdf
Format: Adobe PDF | Size: 13.07 MB
DownloadShow Preview
Thumbnail

Item Export Bar

This item is licensed under a Creative Commons License Creative Commons