Neural Network Surrogates for Free Energy Computation of Complex Chemical Systems
This enables gradient-based free energy methods to incorporate complex and machine-learned collective variables, broadening the scope of biochemistry and materials simulations, but it is incremental as it builds on existing neural network and automatic differentiation techniques.
The paper tackled the bottleneck in free energy computation where existing methods require Jacobians of collective variables, which restricts complex or machine-learned variables, by introducing a neural network surrogate framework that learns variables from coordinates and uses automatic differentiation for Jacobians. On an MgCl2 ion-pairing system, it achieved high accuracy for both simple and complex variables, with Jacobian errors following a near-Gaussian distribution suitable for Gaussian Process Regression pipelines.
Free energy reconstruction methods such as Gaussian Process Regression (GPR) require Jacobians of the collective variables (CVs), a bottleneck that restricts the use of complex or machine-learned CVs. We introduce a neural network surrogate framework that learns CVs directly from Cartesian coordinates and uses automatic differentiation to provide Jacobians, bypassing analytical forms. On an MgCl2 ion-pairing system, our method achieved high accuracy for both a simple distance CV and a complex coordination-number CV. Moreover, Jacobian errors also followed a near-Gaussian distribution, making them suitable for GPR pipelines. This framework enables gradient-based free energy methods to incorporate complex and machine-learned CVs, broadening the scope of biochemistry and materials simulations.