On the Connection Between Diffusion Models and Molecular Dynamics
This work offers an incremental improvement for researchers in computational chemistry by linking diffusion models to molecular dynamics simulations.
The authors tackled the problem of connecting diffusion models to molecular dynamics by providing a simplified mathematical derivation of the relationship between noise and forces, and demonstrated this by training a diffusion-based neural network potential to simulate a coarse-grained lithium chloride solution with enhanced performance through data duplication.
Neural Network Potentials (NNPs) have emerged as a powerful tool for modelling atomic interactions with high accuracy and computational efficiency. Recently, denoising diffusion models have shown promise in NNPs by training networks to remove noise added to stable configurations, eliminating the need for force data during training. In this work, we explore the connection between noise and forces by providing a new, simplified mathematical derivation of their relationship. We also demonstrate how a denoising model can be implemented using a conventional MD software package interfaced with a standard NNP architecture. We demonstrate the approach by training a diffusion-based NNP to simulate a coarse-grained lithium chloride solution and employ data duplication to enhance model performance.