We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Physics > Chemical Physics

Title: Machine learning for molecular simulation

Abstract: 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.
Subjects: Chemical Physics (physics.chem-ph); Machine Learning (cs.LG); Computational Physics (physics.comp-ph); Quantum Physics (quant-ph)
Journal reference: No\'e F, Tkatchenko A, M\"uller KR, Clementi C. 2020. Machine learning for molecular simulation. Annu. Rev. Phys. Chem. 71
DOI: 10.1146/annurev-physchem-042018-052331
Cite as: arXiv:1911.02792 [physics.chem-ph]
  (or arXiv:1911.02792v1 [physics.chem-ph] for this version)

Submission history

From: Frank Noe [view email]
[v1] Thu, 7 Nov 2019 08:14:18 GMT (5884kb,D)

Link back to: arXiv, form interface, contact.