# Adversarial-Residual-Coarse-Graining: Applying machine learning theory to systematic molecular coarse-graining

## TL;DR

This paper introduces a novel machine learning-based framework for systematic molecular coarse-graining using generative adversarial networks, enabling flexible model parameterization including virtual sites, with theoretical and computational validation.

## Contribution

It develops a new coarse-graining approach leveraging GANs, allowing for flexible parameterization and virtual site inclusion, bridging machine learning and molecular modeling.

## Key findings

- Framework can parameterize models with virtual sites.
- Method aligns with Relative Entropy Minimization in ideal cases.
- Demonstrated effectiveness through computational examples.

## Abstract

We utilize connections between molecular coarse-graining approaches and implicit generative models in machine learning to describe a new framework for systematic molecular coarse-graining (CG). Focus is placed on the formalism encompassing generative adversarial networks. The resulting method enables a variety of model parameterization strategies, some of which show similarity to previous CG methods. We demonstrate that the resulting framework can rigorously parameterize CG models containing CG sites with no prescribed connection to the reference atomistic system (termed virtual sites); however, this advantage is offset by the lack of a closed-form expression for the CG Hamiltonian at the resolution obtained after integration over the virtual CG sites. Computational examples are provided for cases in which these methods ideally return identical parameters as Relative Entropy Minimization (REM) CG but where traditional REM CG optimization equations are not applicable.