MLWater - Delta Learning of water potential energy surfaces

In the Project MLWater we explore different techniques for Learning a Neural Network Potential that corrects a classical water model to obtain accurate potential energy surfaces for water molecules in bulk and at interfaces.

The simulation of thermodynamic and structural properties of liquids requires accurate model potentials. In the last decade, machine learning based potentials, including the so called Neural Network (NN) potentials from Behler and Parrinello [1] are gaining influence. Here, the intermolecular potential is constructed from so called Atom-Centered Symmetry Functions that obey translational and rotational symmetry that are fed into a multi-layer Neural Network. Training samples are obtained from quantum mechanical simulations. Our collaborators have recently trained a NN potential for water that recovers experimental thermodynamic properties with high accuracy indicating a high quality result [2].

In this project, we jointly continue this investigation under two aspects: (a) How can an NN model be combined with a traditional water model? (b) How can the training be implemented in a differentiable framework so that experimental data can be taken into account? In question (a) we would like to learn only the difference between a classical force field and quantum mechanical simulations. This would allow us to add new atom types on the force field level without retraining and make sure the electrostatic force is treated on the classical level including its long range part. For question (b) would like to change our training and simulation approach to unlock the potential of differentiable simulation [3]. Recently, two very interesting packages that combine training and execution were released: TorchMD and Jax, MD [4,5]. In this project we will provide a first implementation of a training in each framework as a starting point for future developments.

Forschungszentrum Jülich
  1. Behler, Jörg; Parrinello, Michele. Generalized neural-network representation of high-dimensional potential-energy surfaces. Physical review letters, 2007, 98. Jg., Nr. 14, S. 146401.
  2. Wohlfahrt, Oliver; Dellago, Christoph; Sega, Marcello. Ab initio structure and thermodynamics of the RPBE-D3 water/vapor interface by neural-network molecular dynamics. The Journal of Chemical Physics, 2020, 153. Jg., Nr. 14, S. 144710.
  3. Wang, Wujie; Axelrod, Simon; Gomez-Bombarelli, Rafael. Differentiable molecular simulations for control and learning. arXiv preprint arXiv:2003.00868, 2020.
  4. https://github.com/wwang2/torchmd
  5. https://github.com/google/jax-md
Last Modified: 30.08.2023