Self-learning with neural networks in determinant quantum Monte Carlo studies of the Holstein model.

ORAL

Abstract

Machine learning techniques have recently occupied the focus of many investigators in computational many-body physics. In particular, some practitioners of quantum Monte-Carlo have considered the efficacy of various "Self-Learning" techniques which aim to reduce CPU runtime associated with updates and autocorrelation times. We have used artificial neural networks (NN) within determinant quantum Monte-Carlo to improve the scaling of CPU runtime with typical system parameters. This work focuses on a singleband Holstein Hamiltonian, which models Einstein phonons coupled to on-site electrons. We have implemented both fully connected and convolutional NN and used them to study the metallic and insulating phases of this model. To close, we will assess the generality of this approach to other model systems.

–

Presenters

  • Philip Dee

    University of Tennessee

Authors

  • Shaozhi Li

    Department of Physics and Astronomy, University of Michigan, Physics, University of Michigan

  • Philip Dee

    University of Tennessee

  • Ehsan Khatami

    Department of Physics and Astronomy, San Jose State Unversity, San Jose State University, Physics, San Jose State University

  • Steven Johnston

    Department of Physics and Astronomy, Univ of Tennessee, Knoxville, Department of Physics and Astronomy, University of Tennesse, Physics and Astronomy, University of Tennessee, University of Tennessee, Department of Physics and Astronomy, University of Tennessee, Department of Physics and Astronomy, University of Tennessee, Knoxville