Project

Interacting Particle Networks: a new deep learning approach to molecular simulation of condensed phases.

Code
3F020717
Duration
01 October 2017 → 30 September 2021
Funding
Research Foundation - Flanders (FWO)
Research disciplines
  • Natural sciences
    • Machine learning and decision making
    • High performance computing
    • Modelling and simulation
    • Numerical computation
    • Molecular physics
    • Classical mechanics
    • Classical statics
    • Electrostatics
    • Statistical physics
    • Thermodynamics
    • Classical physics not elsewhere classified
    • Crystallography
    • Dielectrics, piezoelectrics and ferroelectrics
    • Electronic (transport) properties
    • Nanophysics and nanosystems
    • Structural and mechanical properties
    • Phase transformations
    • Thermodynamics
    • Statistical mechanics
    • Quantum physics not elsewhere classified
    • Applied and interdisciplinary physics
    • Computational physics
    • Theory and design of materials
    • Cheminformatics
    • Quantum chemistry
    • Statistical mechanics in chemistry
    • Theoretical and computational chemistry not elsewhere classified
  • Medical and health sciences
    • Biomolecular modelling and design
  • Engineering and technology
    • High performance computing
    • Modelling and simulation
    • Numerical computation
Keywords
Molecular simulations Machine learning Force fields
 
Project description

Force fields are computationally very efficient, yet coarse approximations to the potential energy surface felt by nuclei in molecules. In this project, recent breakthroughs in machine learning will be exploited to increase their reliability. The goal of this work, is to establish force fields with a novel deep learning concept, designed to “understand” many-body interactions: the Interacting Particle Network (IPN).