Stochastic Gradient Approximation: An Efficient Method to Optimize Many-Body Wave Functions

Phys. Rev. Lett. 79, 1173 – Published 18 August 1997
A. Harju, B. Barbiellini, S. Siljamäki, R. M. Nieminen, and G. Ortiz

Abstract

A novel, efficient optimization method for physical problems is presented. The method utilizes the noise inherent in stochastic functions. As an application, an algorithm for the variational optimization of quantum many-body wave functions is derived. The numerical results indicate superior performance when compared to traditional techniques.

DOI: http://dx.doi.org/10.1103/PhysRevLett.79.1173

  • Received 13 September 1996
  • Published in the issue dated 18 August 1997

© 1997 The American Physical Society

Authors & Affiliations

A. Harju, B. Barbiellini, S. Siljamäki, and R. M. Nieminen

  • Laboratory of Physics, Helsinki University of Technology, FIN-02150 Espoo, Finland

G. Ortiz

  • Department of Physics, University of Illinois at Urbana-Champaign, 1110 West Green Street, Urbana, Illinois 61801

References (Subscription Required)

Authorization Required


×
×

Images

×

Log In

Cancel
×

Search


Article Lookup
Paste a citation or DOI

Enter a citation
×
  1. Enter a citation to look up or terms to search.

    Ex: "PRL 112 068103", "Phys. Rev. Lett. 112, 068103", "10.1103/PhysRevLett.112.068103"