Accelerating Molecular Dynamics Simulations with GPU and Machine Learning

EasyChair Preprint 13991

Abstract

The field of molecular dynamics (MD) simulations has undergone significant transformation with the advent of advanced computational techniques, notably the integration of Graphics Processing Units (GPUs) and machine learning (ML). This paper explores the synergy between GPU acceleration and ML algorithms to enhance the efficiency and accuracy of MD simulations. GPUs, with their massive parallel processing capabilities, have revolutionized computational chemistry by dramatically reducing the time required for simulations. Machine learning, on the other hand, offers sophisticated methods for predicting molecular behavior and optimizing simulation parameters. By combining these technologies, we achieve unprecedented simulation speeds and predictive accuracy, enabling more detailed and extensive studies of molecular systems. This integration not only accelerates the exploration of complex biochemical processes but also facilitates real-time simulations, opening new avenues in drug discovery, materials science, and molecular engineering. Our findings demonstrate that the convergence of GPU and ML technologies significantly enhances the performance of MD simulations, paving the way for groundbreaking advancements in computational molecular science.

Keyphrases: Graphics Processing Units, machine learning, molecular dynamics

@booklet{EasyChair:13991,
  author    = {Abi Cit},
  title     = {Accelerating Molecular Dynamics Simulations with GPU and Machine Learning},
  howpublished = {EasyChair Preprint 13991},
  year      = {EasyChair, 2024}}