Neural networks-based variationally enhanced sampling
- Department of Physics, ETH Zurich, 8092 Zurich, Switzerland and Facoltà di Informatica, Instituto di Scienze Computazionali, Università della Svizzera italiana, 6900 Lugano, Switzerland
- Department of Chemistry and Applied Biosciences, ETH Zurich, 8092 Zurich, Switzerland and Facoltà di Informatica, Instituto di Scienze Computazionali, Università della Svizzera italiana (USI), 6900 Lugano, Switzerland
DOI10.24435/materialscloud:2019.0065/v1 (version v1, submitted on 22 October 2019)
How to cite this entry
Luigi Bonati, Yue-Yu Zhang, Michele Parrinello, Neural networks-based variationally enhanced sampling, Materials Cloud Archive (2019), doi: 10.24435/materialscloud:2019.0065/v1.
Sampling complex free-energy surfaces is one of the main challenges of modern atomistic simulation methods. The presence of kinetic bottlenecks in such surfaces often renders a direct approach useless. A popular strategy is to identify a small number of key collective variables and to introduce a bias potential that is able to favor their fluctuations in order to accelerate sampling. Here, we propose to use machine-learning techniques in conjunction with the recent variationally enhanced sampling method [O. Valsson, M. Parrinello, Phys. Rev. Lett. 113, 090601 (2014)] in order to determine such potential. This is achieved by expressing the bias as a neural network. The parameters are determined in a variational learning scheme aimed at minimizing an appropriate functional. This required the development of a more efficient minimization technique. The expressivity of neural networks allows representing rapidly varying free-energy surfaces, removes boundary effects artifacts, and allows several collective variables to be handled.
Materials Cloud sections using this data
No Explore or Discover sections associated with this archive entry.
|42.7 MiB||Inputs and results (ZIP)|
22 October 2019 [This version]