Probing Accuracy-Speedup Tradeoff in Machine Learning Surrogates for Molecular Dynamics Simulations

Sun, Fanbo; Kadupitiya, JCS; Jadhao, Vikram

doi:10.1021/acs.jctc.2c01282

Probing Accuracy-Speedup Tradeoff in Machine Learning Surrogates for Molecular Dynamics Simulations

Journal Article · Mon Apr 24 00:00:00 EDT 2023 · Journal of Chemical Theory and Computation

DOI:https://doi.org/10.1021/acs.jctc.2c01282· OSTI ID:3000914

Sun, Fanbo ^[1]; Kadupitiya, JCS ^[1]; ^[1]

Indiana University, Bloomington, IN (United States)

The performance promise of machine learning surrogates of molecular dynamics simulations of soft materials is significant but generally comes at the cost of acquiring large training datasets to learn the complex relationships between input soft material attributes and output properties. Under the constraint of limited high-performance computing resources, optimizing the size of the training datasets becomes paramount. Using an artificial neural network based surrogate for molecular dynamics simulations of confined electrolytes, we explore the tradeoff between surrogate accuracy and computational gains. Accuracy is assessed by computing the root-mean-square errors between the surrogate predictions and the ground truth results obtained via molecular dynamics simulations. The computational performance is judged by evaluating the speedup which incorporates the training dataset creation time. Improvement in accuracy occurs with a loss of speedup, which scales as the inverse of the training dataset size. Furthermore, the link between surrogate generalizability and the accuracy-speedup tradeoff is assessed by examining the errors incurred in surrogate predictions on unseen, interpolated input variables and developing a net speedup metric to capture the associated gains.

View Accepted Manuscript (DOE)

Research Organization:: University of Virginia, Charlottesville, VA (United States)

Sponsoring Organization:: National Science Foundation (NSF); USDOE

Grant/Contract Number:: SC0023452

OSTI ID:: 3000914

Alternate ID(s):: OSTI ID: 2422252

Journal Information:: Journal of Chemical Theory and Computation, Journal Name: Journal of Chemical Theory and Computation Journal Issue: 14 Vol. 19; ISSN 1549-9618; ISSN 1549-9626

Publisher:: American Chemical Society (ACS)Copyright Statement

Country of Publication:: United States

Language:: English

References (45)

Machine learning for crystal identification and discovery Spellings, Matthew; Glotzer, Sharon C. AIChE Journal, Vol. 64, Issue 6 https://doi.org/10.1002/aic.16157	journal	March 2018
Fast Parallel Algorithms for Short-Range Molecular Dynamics Plimpton, Steve Journal of Computational Physics, Vol. 117, Issue 1 https://doi.org/10.1006/jcph.1995.1039	journal	March 1995
Machine Learning for Performance Enhancement of Molecular Dynamics Simulations Kadupitiya, Jcs; Fox, Geoffrey C.; Jadhao, Vikram Computational Science – ICCS 2019: 19th International Conference, Faro, Portugal, June 12–14, 2019, Proceedings, Part II, p. 116-130 https://doi.org/10.1007/978-3-030-22741-8_9	book	June 2019
Probing the Rheological Properties of Liquids Under Conditions of Elastohydrodynamic Lubrication Using Simulations and Machine Learning Kadupitiya, J. C. S.; Jadhao, Vikram Tribology Letters, Vol. 69, Issue 3 https://doi.org/10.1007/s11249-021-01457-3	journal	May 2021
39 Dimensionality and sample size considerations in pattern recognition practice Jain, A. K.; Chandrasekaran, Balasubramanian Handbook of Statistics, Vol. 2, p. 835-855 https://doi.org/10.1016/S0169-7161(82)02042-2	book	January 1982
Machine learning surrogates for molecular dynamics simulations of soft materials Kadupitiya, J. C. S.; Sun, Fanbo; Fox, Geoffrey Journal of Computational Science, Vol. 42 https://doi.org/10.1016/j.jocs.2020.101107	journal	April 2020
Strange electrostatics in physics, chemistry, and biology Levin, Yan Physica A: Statistical Mechanics and its Applications, Vol. 352, Issue 1 https://doi.org/10.1016/j.physa.2004.12.033	journal	July 2005
Coupling Molecular Dynamics and Deep Learning to Mine Protein Conformational Space Degiacomi, Matteo T. Structure, Vol. 27, Issue 6 https://doi.org/10.1016/j.str.2019.03.018	journal	June 2019
Predicting the Thermodynamic Stability of Solids Combining Density Functional Theory and Machine Learning Schmidt, Jonathan; Shi, Jingming; Borlido, Pedro Chemistry of Materials, Vol. 29, Issue 12 https://doi.org/10.1021/acs.chemmater.7b00156	journal	May 2017
TorchANI: A Free and Open Source PyTorch-Based Deep Learning Implementation of the ANI Neural Network Potentials Gao, Xiang; Ramezanghorbani, Farhad; Isayev, Olexandr Journal of Chemical Information and Modeling, Vol. 60, Issue 7 https://doi.org/10.1021/acs.jcim.0c00451	journal	June 2020
TorchMD: A Deep Learning Framework for Molecular Simulations Doerr, Stefan; Majewski, Maciej; Pérez, Adrià Journal of Chemical Theory and Computation, Vol. 17, Issue 4 https://doi.org/10.1021/acs.jctc.0c01343	journal	March 2021
Critical Knowledge Gaps in Mass Transport through Single-Digit Nanopores: A Review and Perspective Faucher, Samuel; Aluru, Narayana; Bazant, Martin Z. The Journal of Physical Chemistry C, Vol. 123, Issue 35 https://doi.org/10.1021/acs.jpcc.9b02178	journal	May 2019
The Electrostatic Screening Length in Concentrated Electrolytes Increases with Concentration Smith, Alexander M.; Lee, Alpha A.; Perkin, Susan The Journal of Physical Chemistry Letters, Vol. 7, Issue 12 https://doi.org/10.1021/acs.jpclett.6b00867	journal	May 2016
Molecular Dynamics Properties without the Full Trajectory: A Denoising Autoencoder Network for Properties of Simple Liquids Moradzadeh, Alireza; Aluru, N. R. The Journal of Physical Chemistry Letters, Vol. 10, Issue 24 https://doi.org/10.1021/acs.jpclett.9b02820	journal	November 2019
Machine Learning of Coarse-Grained Molecular Dynamics Force Fields Wang, Jiang; Olsson, Simon; Wehmeyer, Christoph ACS Central Science https://doi.org/10.1021/acscentsci.8b00913	journal	April 2019
Machine Learning Enhanced Computational Reverse Engineering Analysis for Scattering Experiments (CREASE) to Determine Structures in Amphiphilic Polymer Solutions Wessels, Michiel G.; Jayaraman, Arthi ACS Polymers Au, Vol. 1, Issue 3 https://doi.org/10.1021/acspolymersau.1c00015	journal	July 2021
Tuning Transport Properties of Nanofluidic Devices with Local Charge Inversion He, Yan; Gillespie, Dirk; Boda, Dezsö Journal of the American Chemical Society, Vol. 131, Issue 14 https://doi.org/10.1021/ja808717u	journal	April 2009
Materials for next-generation desalination and water purification membranes Werber, Jay R.; Osuji, Chinedum O.; Elimelech, Menachem Nature Reviews Materials, Vol. 1, Issue 5 https://doi.org/10.1038/natrevmats.2016.18	journal	April 2016
Active learning of deep surrogates for PDEs: application to metasurface design Pestourie, Raphaël; Mroueh, Youssef; Nguyen, Thanh V. npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-00431-2	journal	October 2020
The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics Yao, Kun; Herr, John E.; Toth, David W. Chemical Science, Vol. 9, Issue 8 https://doi.org/10.1039/C7SC04934J	journal	January 2018
How machine learning can assist the interpretation of ab initio molecular dynamics simulations and conceptual understanding of chemistry Häse, Florian; Fdez. Galván, Ignacio; Aspuru-Guzik, Alán Chemical Science, Vol. 10, Issue 8 https://doi.org/10.1039/C8SC04516J	journal	January 2019
Deep neural network learning of complex binary sorption equilibria from molecular simulation data Sun, Yangzesheng; DeJaco, Robert F.; Siepmann, J. Ilja Chemical Science, Vol. 10, Issue 16 https://doi.org/10.1039/C8SC05340E	journal	January 2019
Electrostatic potential inside ionic solutions confined by dielectrics: a variational approach Allen, Rosalind; Hansen, Jean-Pierre; Melchionna, Simone Physical Chemistry Chemical Physics, Vol. 3, Issue 19 https://doi.org/10.1039/b105176h	journal	January 2001
An iterative, fast, linear-scaling method for computing induced charges on arbitrary dielectric boundaries Tyagi, Sandeep; Süzen, Mehmet; Sega, Marcello The Journal of Chemical Physics, Vol. 132, Issue 15 https://doi.org/10.1063/1.3376011	journal	April 2010
How to mesh up Ewald sums. I. A theoretical and numerical comparison of various particle mesh routines Deserno, Markus; Holm, Christian The Journal of Chemical Physics, Vol. 109, Issue 18 https://doi.org/10.1063/1.477414	journal	November 1998
Efficient and accurate simulation of dynamic dielectric objects Barros, Kipton; Sinkovits, Daniel; Luijten, Erik The Journal of Chemical Physics, Vol. 140, Issue 6 https://doi.org/10.1063/1.4863451	journal	February 2014
Ionic structure in liquids confined by dielectric interfaces Jing, Yufei; Jadhao, Vikram; Zwanikken, Jos W. The Journal of Chemical Physics, Vol. 143, Issue 19 https://doi.org/10.1063/1.4935704	journal	November 2015
Adaptive enhanced sampling by force-biasing using neural networks Guo, Ashley Z.; Sevgen, Emre; Sidky, Hythem The Journal of Chemical Physics, Vol. 148, Issue 13 https://doi.org/10.1063/1.5020733	journal	April 2018
Ionic structure and decay length in highly concentrated confined electrolytes Anousheh, Nasim; Solis, Francisco J.; Jadhao, Vikram AIP Advances, Vol. 10, Issue 12 https://doi.org/10.1063/5.0028003	journal	December 2020
Relationship between local structure and relaxation in out-of-equilibrium glassy systems Schoenholz, Samuel S.; Cubuk, Ekin D.; Kaxiras, Efthimios Proceedings of the National Academy of Sciences, Vol. 114, Issue 2 https://doi.org/10.1073/pnas.1610204114	journal	December 2016
Machine learning and data science in soft materials engineering Ferguson, Andrew L. Journal of Physics: Condensed Matter, Vol. 30, Issue 4 https://doi.org/10.1088/1361-648X/aa98bd	journal	December 2017
Building high accuracy emulators for scientific simulations with deep neural architecture search Kasim, M. F.; Watson-Parris, D.; Deaconu, L. Machine Learning: Science and Technology, Vol. 3, Issue 1 https://doi.org/10.1088/2632-2153/ac3ffa	journal	December 2021
Solving Newton’s equations of motion with large timesteps using recurrent neural networks based operators Kadupitiya, J. C. S.; Fox, Geoffrey C.; Jadhao, Vikram Machine Learning: Science and Technology, Vol. 3, Issue 2 https://doi.org/10.1088/2632-2153/ac5f60	journal	April 2022
Computing induced charges in inhomogeneous dielectric media: Application in a Monte Carlo simulation of complex ionic systems Boda, Dezsö; Gillespie, Dirk; Nonner, Wolfgang Physical Review E, Vol. 69, Issue 4 https://doi.org/10.1103/PhysRevE.69.046702	journal	April 2004
Correlated electrolyte solutions and ion-induced attractions between nanoparticles Zwanikken, Jos W.; de la Cruz, Monica Olvera Physical Review E, Vol. 82, Issue 5 https://doi.org/10.1103/PhysRevE.82.050401	journal	November 2010
Small sample size effects in statistical pattern recognition: recommendations for practitioners Raudys, S. J.; Jain, A. K. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 13, Issue 3 https://doi.org/10.1109/34.75512	journal	March 1991
Molecular Dynamics Simulations on Cloud Computing and Machine Learning Platforms Sharma, Prateek; Jadhao, Vikram 2021 IEEE 14th International Conference on Cloud Computing (CLOUD) https://doi.org/10.1109/CLOUD53861.2021.00101	conference	September 2021
DeepDriveMD: Deep-Learning Driven Adaptive Molecular Simulations for Protein Folding Lee, Hyungro; Turilli, Matteo; Jha, Shantenu 2019 IEEE/ACM Third Workshop on Deep Learning on Supercomputers (DLS) https://doi.org/10.1109/DLS49591.2019.00007	conference	November 2019
Integrating Machine Learning with HPC-driven Simulations for Enhanced Student Learning Jadhao, Vikram; Kadupitiya, Jcs 2020 IEEE/ACM Workshop on Education for High-Performance Computing (EduHPC) https://doi.org/10.1109/EduHPC51895.2020.00009	conference	November 2020
Maximizing the right stuff: The trade-off between membrane permeability and selectivity Park, Ho Bum; Kamcev, Jovan; Robeson, Lloyd M. Science, Vol. 356, Issue 6343 https://doi.org/10.1126/science.aab0530	journal	June 2017
Methodologic Guide for Evaluating Clinical Performance and Effect of Artificial Intelligence Technology for Medical Diagnosis and Prediction Park, Seong Ho; Han, Kyunghwa Radiology, Vol. 286, Issue 3 https://doi.org/10.1148/radiol.2017171920	journal	March 2018
Machine learning for parameter auto-tuning in molecular dynamics simulations: Efficient dynamics of ions near polarizable nanoparticles Kadupitiya, Jcs; Fox, Geoffrey C.; Jadhao, Vikram The International Journal of High Performance Computing Applications, Vol. 34, Issue 3 https://doi.org/10.1177/1094342019899457	journal	January 2020
AI-driven multiscale simulations illuminate mechanisms of SARS-CoV-2 spike dynamics Casalino, Lorenzo; Dommer, Abigail C.; Gaieb, Zied The International Journal of High Performance Computing Applications https://doi.org/10.1177/10943420211006452	journal	April 2021
HOOMD-TF: GPU-Accelerated, Online Machine Learning in the HOOMD-blue Molecular Dynamics Engine Barrett, Rainier; Chakraborty, Maghesree; Amirkulova, Dilnoza Journal of Open Source Software, Vol. 5, Issue 51 https://doi.org/10.21105/joss.02367	journal	July 2020
Detecting Neuroimaging Biomarkers for Psychiatric Disorders: Sample Size Matters Schnack, Hugo G.; Kahn, René S. Frontiers in Psychiatry, Vol. 7 https://doi.org/10.3389/fpsyt.2016.00050	journal	March 2016

Similar Records

A physics-informed and hierarchically regularized data-driven model for predicting fluid flow through porous media

Journal Article · Thu Oct 14 20:00:00 EDT 2021 · Journal of Computational Physics · OSTI ID:1804349

Transfer learning of neural surrogates on multifidelity groundwater simulations

Dataset · Tue Dec 31 23:00:00 EST 2024 · OSTI ID:3005558

A physics-constrained neural ordinary differential equations approach for robust learning of stiff chemical kinetics

Journal Article · Mon Mar 24 20:00:00 EDT 2025 · Combustion Theory and Modelling · OSTI ID:2588541

Related Subjects

Anions
Cations
Electrical properties
Electrolytes
Interfaces

Probing Accuracy-Speedup Tradeoff in Machine Learning Surrogates for Molecular Dynamics Simulations

Citation Formats

References (45)

Similar Records

Related Subjects