Machine learning for molecular simulation
This is an incremental review paper that addresses the problem of improving efficiency and accuracy in molecular simulations for researchers in computational chemistry and physics.
The paper reviews how machine learning, especially deep neural networks, is applied to molecular simulation to tackle complex calculations like predicting quantum-mechanical energies, coarse-grained dynamics, and free energy extraction, highlighting its transformative impact without specifying new results or numbers.
Machine learning (ML) is transforming all areas of science. The complex and time-consuming calculations in molecular simulations are particularly suitable for a machine learning revolution and have already been profoundly impacted by the application of existing ML methods. Here we review recent ML methods for molecular simulation, with particular focus on (deep) neural networks for the prediction of quantum-mechanical energies and forces, coarse-grained molecular dynamics, the extraction of free energy surfaces and kinetics and generative network approaches to sample molecular equilibrium structures and compute thermodynamics. To explain these methods and illustrate open methodological problems, we review some important principles of molecular physics and describe how they can be incorporated into machine learning structures. Finally, we identify and describe a list of open challenges for the interface between ML and molecular simulation.