Berlin 2018 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
DY: Fachverband Dynamik und Statistische Physik
DY 31: Condensed Matter Simulations augmented by Advanced Statistical Methodologies II (joint session DY/CPP)
DY 31.1: Talk
Tuesday, March 13, 2018, 14:00–14:15, BH-N 128
Nonlinear Network description for many-body quantum systems in continuous space — •Markus Holzmann1, Michele Ruggeri2, and Saverio Moroni3 — 1LPMMC, UMR 5493 of CNRS, Université Grenoble Alpes, F-38100 Grenoble France — 2Max Planck Institute for Solid State Research, Heisenbergstr. 1, 70569 Stuttgart, Germany — 3DEMOCRITOS National Simulation Center, Istituto Officina dei Materiali del CNR and SISSA, Via Bonomea 265, I-34136 Trieste, Italy
We show that the recently introduced iterative backflow renormalization [1] can be interpreted as a general neural network in continuum space with non-linear functions in the hidden units [2]. We use this wave function within Variational Monte Carlo for fermionic and bosonic liquid helium in two and three dimensions, where we typically find a tenfold increase in accuracy over currently used wave functions. For two dimensional 4 He, we also show that the iterative backflow wave function can describe both the liquid and the solid phase with the same functional form - a feature shared with the Shadow Wave Function, but now joined by much higher accuracy.
[1] M. Taddei, M. Ruggeri, S. Moroni, and M. Holzmann, Phys. Rev. B 91, 115106 (2015).
[2] M. Ruggeri, S. Moroni, and M. Holzmann, cond-mat/1711.01993 (2017).