Home > Research > Publications & Outputs > Discovering quantum phase transitions with ferm...

Research

Associated organisational unit

Electronic data

  • Electron_Gas_Paper

    Accepted author manuscript, 1.6 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

View graph of relations

Discovering quantum phase transitions with fermionic neural networks

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Discovering quantum phase transitions with fermionic neural networks. / Cassella, Gino; Sutterud, Halvard; Azadi, Sam et al.
In: Physical review letters, Vol. 130, No. 3, 036401, 20.01.2023.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Cassella, G, Sutterud, H, Azadi, S, Drummond, N, Pfau, D, Spencer, JS & Foulkes, WMC 2023, 'Discovering quantum phase transitions with fermionic neural networks', Physical review letters, vol. 130, no. 3, 036401. <https://doi.org/10.1103/PhysRevLett.130.036401>

APA

Cassella, G., Sutterud, H., Azadi, S., Drummond, N., Pfau, D., Spencer, J. S., & Foulkes, W. M. C. (2023). Discovering quantum phase transitions with fermionic neural networks. Physical review letters, 130(3), Article 036401. https://doi.org/10.1103/PhysRevLett.130.036401

Vancouver

Cassella G, Sutterud H, Azadi S, Drummond N, Pfau D, Spencer JS et al. Discovering quantum phase transitions with fermionic neural networks. Physical review letters. 2023 Jan 20;130(3):036401. Epub 2023 Jan 20.

Author

Cassella, Gino ; Sutterud, Halvard ; Azadi, Sam et al. / Discovering quantum phase transitions with fermionic neural networks. In: Physical review letters. 2023 ; Vol. 130, No. 3.

Bibtex

@article{a48a319500054d8e9701ae9e89b63c87,
title = "Discovering quantum phase transitions with fermionic neural networks",
abstract = "Deep neural networks have been very successful as highly accurate wave function Ans{\"a}tze for variational Monte Carlo calculations of molecular ground states. We present an extension of one such Ansatz, FermiNet, to calculations of the ground states of periodic Hamiltonians, and study the homogeneous electron gas. FermiNet calculations of the ground-state energies of small electron gas systems are in excellent agreement with previous initiator full configuration interaction quantum Monte Carlo and diffusion Monte Carlo calculations. We investigate the spin-polarized homogeneous electron gas and demonstrate that the same neural network architecture is capable of accurately representing both the delocalized Fermi liquid state and the localized Wigner crystal state. The network converges on the translationally invariant ground state at high density and spontaneously breaks the symmetry to produce the crystalline ground state at low density, despite being given no a priori knowledge that a phase transition exists.",
author = "Gino Cassella and Halvard Sutterud and Sam Azadi and Neil Drummond and David Pfau and Spencer, {James S.} and Foulkes, {W. M. C.}",
year = "2023",
month = jan,
day = "20",
language = "English",
volume = "130",
journal = "Physical review letters",
issn = "1079-7114",
publisher = "American Physical Society",
number = "3",

}

RIS

TY - JOUR

T1 - Discovering quantum phase transitions with fermionic neural networks

AU - Cassella, Gino

AU - Sutterud, Halvard

AU - Azadi, Sam

AU - Drummond, Neil

AU - Pfau, David

AU - Spencer, James S.

AU - Foulkes, W. M. C.

PY - 2023/1/20

Y1 - 2023/1/20

N2 - Deep neural networks have been very successful as highly accurate wave function Ansätze for variational Monte Carlo calculations of molecular ground states. We present an extension of one such Ansatz, FermiNet, to calculations of the ground states of periodic Hamiltonians, and study the homogeneous electron gas. FermiNet calculations of the ground-state energies of small electron gas systems are in excellent agreement with previous initiator full configuration interaction quantum Monte Carlo and diffusion Monte Carlo calculations. We investigate the spin-polarized homogeneous electron gas and demonstrate that the same neural network architecture is capable of accurately representing both the delocalized Fermi liquid state and the localized Wigner crystal state. The network converges on the translationally invariant ground state at high density and spontaneously breaks the symmetry to produce the crystalline ground state at low density, despite being given no a priori knowledge that a phase transition exists.

AB - Deep neural networks have been very successful as highly accurate wave function Ansätze for variational Monte Carlo calculations of molecular ground states. We present an extension of one such Ansatz, FermiNet, to calculations of the ground states of periodic Hamiltonians, and study the homogeneous electron gas. FermiNet calculations of the ground-state energies of small electron gas systems are in excellent agreement with previous initiator full configuration interaction quantum Monte Carlo and diffusion Monte Carlo calculations. We investigate the spin-polarized homogeneous electron gas and demonstrate that the same neural network architecture is capable of accurately representing both the delocalized Fermi liquid state and the localized Wigner crystal state. The network converges on the translationally invariant ground state at high density and spontaneously breaks the symmetry to produce the crystalline ground state at low density, despite being given no a priori knowledge that a phase transition exists.

M3 - Journal article

VL - 130

JO - Physical review letters

JF - Physical review letters

SN - 1079-7114

IS - 3

M1 - 036401

ER -