Search papers, labs, and topics across Lattice.
The paper introduces a method to improve the thermodynamic transferability of machine-learned coarse-grained (ML CG) force fields by training on the thermal response forces of the potential of mean force (PMF). By incorporating these thermal response terms into the ML CG FFs, the resulting models exhibit significantly improved transferability across different temperatures. The method is demonstrated on CG water models, showing accurate and predictive CG dynamics at temperatures different from the training temperature.
Temperature-dependent coarse-grained models, notoriously hard to train, become far more transferable thanks to a new method that learns thermal response forces.
Machine-learned (ML) coarse-grained (CG) models are a promising tool for significantly enhancing the efficiency of molecular simulations by systematically removing degrees of freedom while retaining fidelity to the underlying fine-grained model. The CG potential of mean force (PMF) is inherently dependent on thermodynamic conditions and, hence, a CG force-field (FF) which is trained at one thermodynamic state point is not necessarily accurate at another. We propose, in this work, a novel and data-efficient means of learning temperature dependence into ML CG force-fields via training on the thermal response forces of the PMF. We demonstrate how incorporating these terms into ML CG FFs confers significantly improved transferability for CG water models and demonstrate how this transferability enables accurate and predictive CG dynamics.
