pith. sign in

arxiv: 2606.12928 · v1 · pith:BRF6KL4Ynew · submitted 2026-06-11 · ❄️ cond-mat.quant-gas · cond-mat.dis-nn· cond-mat.str-el· physics.comp-ph· quant-ph

Continuum Neural Momentum Eigenstate for Variationally Solving Quasiparticles

David D. Dai , Marin Solja\v{c}i\'c This is my paper

Pith reviewed 2026-06-27 05:22 UTC · model grok-4.3

classification ❄️ cond-mat.quant-gas cond-mat.dis-nncond-mat.str-elphysics.comp-phquant-ph
keywords neural quantum statesvariational Monte Carlomomentum eigenstatesquasiparticles2D bosonsroton minimumsuperfluidWigner crystal
0
0 comments X

The pith

A neural quantum state is constructed to be an exact eigenstate of total momentum for any chosen value, allowing standard variational Monte Carlo to target ground states in fixed momentum sectors.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents an architecture for neural quantum states in continuum systems that guarantees the wavefunction is an exact eigenstate of total momentum with any selected eigenvalue. This property lets off-the-shelf variational Monte Carlo optimize directly inside a chosen momentum sector without additional projection steps. When tested on two-dimensional bosons with 1/r interactions, the same ansatz form produces dispersion relations whose shapes identify superfluid, roton, crystal, and phonon regimes at different densities. Density-density correlations are used to verify the phase assignments and to examine the internal structure of the excitations, including unexpected multi-particle phase strings in the roton state.

Core claim

The EVE architecture constructs a neural quantum state that is by design an exact eigenstate of the total momentum operator with eigenvalue equal to any chosen allowed momentum vector k. This property permits standard variational Monte Carlo procedures to variationally minimize the energy within a fixed momentum sector. When applied to two-dimensional bosons with mutual 1/r interactions, the same ansatz form describes four distinct phases—superfluid, roton, crystal, and phonon—across a range of densities, with phase identification following from the shape of the computed dispersion relation and confirmed by correlation functions.

What carries the argument

The EVE neural quantum state architecture, which enforces the exact momentum eigenstate property by construction for any selected total momentum k.

If this is right

  • A single unified variational ansatz can describe four qualitatively different quantum phases: superfluid, roton, crystal, and phonon.
  • The phase of matter can be identified directly from the shape of the dispersion relation extracted at different densities.
  • Density-density correlation functions confirm the phase diagnoses and reveal the correlation structure of the excitations.
  • The roton minimum exhibits multi-particle phase strings formed when several vortex dipoles merge.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The construction could be combined with additional symmetry constraints to target states carrying other quantum numbers in the same systems.
  • The approach may allow mapping of quasiparticle dispersions across wider ranges of density and interaction strength than previously accessible with continuum neural states.
  • Extension to three dimensions or to fermionic statistics would test whether the momentum-eigenstate property remains equally effective.

Load-bearing premise

Variational Monte Carlo optimization of the EVE ansatz converges to the true ground state in each momentum sector without the network form introducing a bias that prevents access to the reported phases.

What would settle it

An independent calculation at the same densities and interaction strengths that produces lower variational energies than those obtained with EVE would show that the ansatz has not reached the ground state.

Figures

Figures reproduced from arXiv: 2606.12928 by David D. Dai, Marin Solja\v{c}i\'c.

Figure 1
Figure 1. Figure 1: FIG. 1. Solid lines show the edge-to-vertex aggregation up [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Top Left: The superfluid’s momentum-sector energies. A pronounced roton minimum exists near [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. All panels show a density-density correlation function, where the conditioning particle is placed at the origin. Top Left: [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Left: Learned phase of [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Comparison of EVE to a neural-Jastrow and Nosanow-neural-Jastrow (only for the crystal) for the superfluid and [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
read the original abstract

We design the first neural quantum state for continuum particles that, for any chosen allowed momentum $\mathbf{k}$, is by construction an exact eigenstate of total momentum with eigenvalue $\mathbf{k}$. Our architecture, EVE, enables off-the-shelf VMC to solve for momentum-sector ground states. We test EVE on 2D bosons with mutual $1/r$ interactions, finding that a single unified ansatz is capable of describing four qualitatively different states: superfluid, roton, crystal, and phonon. At different densities, we extract the underlying phase of matter from the dispersion's shape. At $r_s = 20.0$, we see the roton minimum at finite $k$ expected of a superfluid. At $r_s = 100.0$, we see striking zone folding indicative of crystalline order, with periodically spaced minima representing floating crystals connected by phonon arcs in between. Using density-density correlation functions, we confirm the phase diagnoses and probe the excitations' correlation structures. Finally, we analyze the roton's phase texture and find unexpected multi-particle phase strings, formed when several vortex dipoles merge, leaving two vortices connected by a phase slip.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript introduces the EVE neural quantum state architecture for continuum particles that is, by construction, an exact eigenstate of the total momentum operator for any chosen allowed momentum k. This enables off-the-shelf variational Monte Carlo (VMC) to target momentum-sector ground states. Applied to 2D bosons with mutual 1/r interactions, a single unified ansatz is shown to capture four qualitatively different states—superfluid (roton minimum at finite k for rs=20), crystal (zone folding with periodic minima at rs=100), and phonon—diagnosed from dispersion shapes, confirmed via density-density correlation functions, and further analyzed through the roton's phase texture revealing multi-particle phase strings.

Significance. The exact momentum eigenstate construction by design is a clear technical strength that directly supports dispersion calculations without additional symmetry enforcement. If the VMC optimizations reach the true ground states, the unified ansatz across phases offers a practical route to quasiparticle studies in long-range interacting continuum systems. The work demonstrates qualitative access to multiple phases with one network, which could aid exploration of exotic states, though quantitative validation against benchmarks would strengthen its utility.

major comments (1)
  1. [Results (rs=20 and rs=100 cases)] The central claim that a single EVE ansatz plus off-the-shelf VMC solves for the true momentum-sector ground states (enabling the reported phase identifications) is load-bearing but rests on an unverified assumption of convergence without representational bias. For the crystal phase at rs=100 and roton features at rs=20, the risk of metastable trapping or bias against broken translational symmetry in long-range 1/r systems is not addressed by the described evidence; explicit multi-seed energy comparisons or small-system benchmarks against exact methods are required to substantiate that the observed dispersion shapes and correlations reflect ground states rather than ansatz artifacts.
minor comments (2)
  1. [Abstract and Numerical Results] The abstract and results would benefit from reporting particle number N, system size, and any convergence metrics (e.g., energy variance or statistical error bars on the dispersion) to allow assessment of the quantitative reliability of the phase diagnoses.
  2. [Introduction/Methods] Notation for the momentum k and the interaction parameter rs should be defined at first use with explicit reference to the Hamiltonian to improve clarity for readers outside the immediate subfield.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and constructive criticism. We agree that stronger evidence of convergence is needed to support the central claims and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Results (rs=20 and rs=100 cases)] The central claim that a single EVE ansatz plus off-the-shelf VMC solves for the true momentum-sector ground states (enabling the reported phase identifications) is load-bearing but rests on an unverified assumption of convergence without representational bias. For the crystal phase at rs=100 and roton features at rs=20, the risk of metastable trapping or bias against broken translational symmetry in long-range 1/r systems is not addressed by the described evidence; explicit multi-seed energy comparisons or small-system benchmarks against exact methods are required to substantiate that the observed dispersion shapes and correlations reflect ground states rather than ansatz artifacts.

    Authors: We acknowledge that the original manuscript lacks explicit multi-seed statistics and small-system exact benchmarks, which leaves open the possibility of metastable trapping or representational bias in long-range 1/r systems. Our optimizations used standard VMC with random initializations of network parameters and selected the lowest-energy runs, but we did not report seed-to-seed variance or direct comparisons to exact methods. The qualitative features (roton minimum at rs=20, zone folding at rs=100) together with density-density correlations provide supporting evidence, yet we agree these are indirect. In the revised manuscript we will add: (i) energy distributions and dispersion consistency across at least five independent random seeds for representative k-points at both rs values, and (ii) for small particle numbers (N=4,6) where exact diagonalization is feasible, direct energy and correlation comparisons in the relevant momentum sectors. These additions will allow readers to assess convergence and any bias against broken translational symmetry. We therefore accept the referee's request for additional validation. revision: yes

Circularity Check

0 steps flagged

No significant circularity; momentum eigenstate property is explicitly constructed rather than derived

full rationale

The paper explicitly designs the EVE architecture so that the neural quantum state is an exact momentum eigenstate by construction for any chosen k, then applies standard VMC to variationally optimize within each sector. The reported phase diagnoses (roton minimum, zone folding, phonon arcs) are obtained by inspecting the resulting dispersion and density-density correlations from the optimized states. No load-bearing step reduces a claimed prediction to a fitted input, self-citation chain, or ansatz smuggled via prior work; the symmetry enforcement is openly stated as a design choice rather than presented as a novel derived result. The derivation chain therefore remains self-contained as a variational method without circular reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract alone supplies insufficient detail to enumerate free parameters, axioms, or invented entities. The 1/r interaction and chosen densities (rs = 20, 100) are inputs from the problem setup rather than fitted quantities introduced by the method.

pith-pipeline@v0.9.1-grok · 5753 in / 1321 out tokens · 18842 ms · 2026-06-27T05:22:42.209235+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

75 extracted references · 9 canonical work pages · 3 internal anchors

  1. [1]

    Landau,Theoryofthesuperfluidityofhelium II,Phys

    L. Landau,Theoryofthesuperfluidityofhelium II,Phys. Rev.60, 356 (1941)

    1941

  2. [2]

    R. P. Feynman, Atomic theory of the two-fluid model of liquid helium, Phys. Rev.94, 262 (1954)

    1954

  3. [3]

    R. P. Feynman and M. Cohen, Energy spectrum of the excitations in liquid helium, Phys. Rev.102, 1189 (1956)

    1956

  4. [4]

    Nambu, Quasi-particles and gauge invariance in the theory of superconductivity, Phys

    Y. Nambu, Quasi-particles and gauge invariance in the theory of superconductivity, Phys. Rev.117, 648 (1960)

    1960

  5. [5]

    Goldstone, Field theories with «superconductor» solu- tions, Il Nuovo Cimento19, 154–164 (1961)

    J. Goldstone, Field theories with «superconductor» solu- tions, Il Nuovo Cimento19, 154–164 (1961)

    1961

  6. [6]

    Goldstone, A

    J. Goldstone, A. Salam, and S. Weinberg, Broken sym- metries, Phys. Rev.127, 965 (1962)

    1962

  7. [7]

    D. C. Tsui, H. L. Stormer, and A. C. Gossard, Two- dimensional magnetotransport in the extreme quantum limit, Phys. Rev. Lett.48, 1559 (1982)

    1982

  8. [8]

    R. B. Laughlin, Anomalous quantum hall effect: An in- compressible quantum fluid with fractionally charged ex- citations, Phys. Rev. Lett.50, 1395 (1983)

    1983

  9. [9]

    Arovas, J

    D. Arovas, J. R. Schrieffer, and F. Wilczek, Fractional statistics and the quantum hall effect, Phys. Rev. Lett. 53, 722 (1984)

    1984

  10. [10]

    Moore and N

    G. Moore and N. Read, Nonabelions in the fractional quantum hall effect, Nuclear Physics B360, 362 (1991)

    1991

  11. [11]

    X. G. Wen, Topological orders in rigid states, Interna- tional Journal of Modern Physics B04, 239 (1990)

    1990

  12. [12]

    X. G. Wen and Q. Niu, Ground-state degeneracy of the fractional quantum hall states in the presence of a ran- dom potential and on high-genus riemann surfaces, Phys. Rev. B41, 9377 (1990)

    1990

  13. [13]

    W. L. McMillan, Ground state of liquidHe4, Phys. Rev. 138, A442 (1965)

    1965

  14. [14]

    Ceperley, G

    D. Ceperley, G. V. Chester, and M. H. Kalos, Monte Carlo simulation of a many-fermion study, Phys. Rev. B 16, 3081 (1977)

    1977

  15. [15]

    D. M. Ceperley and B. J. Alder, Ground state of the electron gas by a stochastic method, Phys. Rev. Lett. 45, 566 (1980)

    1980

  16. [16]

    Tanatar and D

    B. Tanatar and D. M. Ceperley, Ground state of the two- 9 dimensional electron gas, Phys. Rev. B39, 5005 (1989)

    1989

  17. [17]

    W. M. C. Foulkes, L. Mitas, R. J. Needs, and G. Ra- jagopal, Quantum Monte Carlo simulations of solids, Rev. Mod. Phys.73, 33 (2001)

    2001

  18. [18]

    R. J. Needs, M. D. Towler, N. D. Drummond, and P.LópezRíos,Continuumvariationalanddiffusionquan- tum Monte Carlo calculations, Journal of Physics: Con- densed Matter22, 023201 (2009)

    2009

  19. [19]

    Zhao and E

    L. Zhao and E. Neuscamman, An efficient variational principle for the direct optimization of excited states, Journal of Chemical Theory and Computation12, 3436–3440 (2016)

    2016

  20. [20]

    K. Choo, G. Carleo, N. Regnault, and T. Neupert, Sym- metries and many-body excitations with neural-network quantum states, Phys. Rev. Lett.121, 167204 (2018)

    2018

  21. [21]

    Pathak, B

    S. Pathak, B. Busemeyer, J. N. B. Rodrigues, and L. K. Wagner, Excited states in variational Monte Carlo using a penalty method, The Journal of Chemical Physics154, 034101 (2021)

    2021

  22. [22]

    M. T. Entwistle, Z. Schätzle, P. A. Erdman, J. Hermann, and F. Noé, Electronic excited states in deep variational Monte Carlo, Nature Communications14, 274 (2023)

    2023

  23. [23]

    W. A. Wheeler, K. G. Kleiner, and L. K. Wagner, En- semble variational Monte Carlo for optimization of corre- lated excited state wave functions, Electronic Structure 6, 025001 (2024)

    2024

  24. [24]

    D. Pfau, S. Axelrod, H. Sutterud, I. von Glehn, and J. S. Spencer, Accurate computation of quantum ex- cited states with neural networks, Science385, eadn0137 (2024)

    2024

  25. [25]

    Lange, F

    H. Lange, F. Döschl, J. Carrasquilla, and A. Bohrdt, Neural network approach to quasiparticle dispersions in doped antiferromagnets, Communications Physics7, 10.1038/s42005-024-01678-7 (2024)

    work page doi:10.1038/s42005-024-01678-7 2024

  26. [26]

    Romero, J

    I. Romero, J. Nys, and G. Carleo, Spectroscopy of two- dimensional interacting lattice electrons using symmetry- aware neural backflow transformations, Communications Physics8, 10.1038/s42005-025-01955-z (2025)

    work page doi:10.1038/s42005-025-01955-z 2025

  27. [27]

    Yoshioka, W

    N. Yoshioka, W. Mizukami, and F. Nori, Solving quasiparticle band spectra of real solids using neural- network quantum states, Communications Physics4, 10.1038/s42005-021-00609-0 (2021)

    work page doi:10.1038/s42005-021-00609-0 2021

  28. [28]

    L. Fu, Y. Wu, H. Shang, and J. Yang, Transformer- based neural-network quantum state method for elec- tronic band structures of real solids, Journal of Chemical Theory and Computation20, 6218–6226 (2024)

    2024

  29. [29]

    and Luo, D

    L. Zhang and D. Luo, Neural transformer backflow for solving momentum-resolved ground states of strongly correlated materials (2025), arXiv:2509.09275 [cond- mat.str-el]

    work page arXiv 2025

  30. [30]

    Carleo and M

    G. Carleo and M. Troyer, Solving the quantum many- body problem with artificial neural networks, Science 355, 602 (2017)

    2017

  31. [31]

    Nomura, A

    Y. Nomura, A. S. Darmawan, Y. Yamaji, and M. Imada, Restricted Boltzmann machine learning for solving strongly correlated quantum systems, Phys. Rev. B96, 205152 (2017)

    2017

  32. [32]

    Carleo, K

    G. Carleo, K. Choo, D. Hofmann, J. E. Smith, T. Wester- hout, F. Alet, E. J. Davis, S. Efthymiou, I. Glasser, S.-H. Lin, M.Mauri, G.Mazzola, C.B.Mendl, E.vanNieuwen- burg, O. O’Reilly, H. Théveniaut, G. Torlai, F. Vicen- tini, and A. Wietek, NetKet: A machine learning toolkit for many-body quantum systems, SoftwareX10, 100311 (2019)

    2019

  33. [33]

    K. Choo, T. Neupert, and G. Carleo, Two-dimensional frustratedJ 1−J2 model studied with neural network quantum states, Phys. Rev. B100, 125124 (2019)

    2019

  34. [34]

    O.Sharir, Y.Levine, N.Wies, G.Carleo,andA.Shashua, Deep autoregressive models for the efficient variational simulation of many-body quantum systems, Phys. Rev. Lett.124, 020503 (2020)

    2020

  35. [35]

    D. Pfau, J. S. Spencer, A. G. D. G. Matthews, and W. M. C. Foulkes, Ab initio solution of the many-electron Schrödinger equation with deep neural networks, Phys. Rev. Res.2, 033429 (2020)

    2020

  36. [36]

    Hermann, Z

    J. Hermann, Z. Schätzle, and F. Noé, Deep-neural- network solution of the electronic Schrödinger equation, Nature Chemistry12, 891–897 (2020)

    2020

  37. [37]

    J. S. Spencer, D. Pfau, A. Botev, and W. M. C. Foulkes, Better, faster fermionic neural networks (2020), arXiv:2011.07125 [physics.comp-ph]

    work page arXiv 2020

  38. [38]

    L. L. Viteritti, R. Rende, and F. Becca, Transformer vari- ational wave functions for frustrated quantum spin sys- tems, Phys. Rev. Lett.130, 236401 (2023)

    2023

  39. [39]

    von Glehn, J

    I. von Glehn, J. S. Spencer, and D. Pfau, A self-attention ansatz for ab-initio quantum chemistry, inThe Eleventh International Conference on Learning Representations (2023)

    2023

  40. [40]

    Hermann, J

    J. Hermann, J. Spencer, K. Choo, A. Mezzacapo, W. M. Foulkes, D. Pfau, G. Carleo, and F. Noé, Ab initio quan- tum chemistry with neural-network wavefunctions, Na- ture Reviews Chemistry7, 692–709 (2023)

    2023

  41. [41]

    H.Lange, A.VandeWalle, A.Abedinnia,andA.Bohrdt, From architectures to applications: a review of neural quantum states, Quantum Science and Technology9, 040501 (2024)

    2024

  42. [42]

    W. Fu, W. Ren, and J. Chen, Variance extrapola- tion method for neural-network variational Monte Carlo, Machine Learning: Science and Technology5, 015016 (2024)

    2024

  43. [43]

    In other words, we scan the phase diagram by changing the kinetic energy’s coefficient instead of scaling the system area

    To stabilize the overall energy scale in the low-density regime, we choosea i such that the area per particle is alwaysπ. In other words, we scan the phase diagram by changing the kinetic energy’s coefficient instead of scaling the system area

  44. [44]

    Neural Machine Translation by Jointly Learning to Align and Translate

    D. Bahdanau, K. Cho, and Y. Bengio, Neural machine translation by jointly learning to align and translate, CoRRabs/1409.0473(2014)

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  45. [45]

    Effective Approaches to Attention-based Neural Machine Translation

    T. Luong, H. Pham, and C. D. Manning, Effective ap- proaches to attention-based neural machine translation, ArXivabs/1508.04025(2015)

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  46. [46]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, Attention is all you need, inAdvances in Neural Infor- mation Processing Systems, Vol. 30, edited by I. Guyon, U.V.Luxburg, S.Bengio, H.Wallach, R.Fergus, S.Vish- wanathan, and R. Garnett (Curran Associates, Inc., 2017)

    2017

  47. [47]

    Gaussian Error Linear Units (GELUs)

    D. Hendrycks and K. Gimpel, Gaussian error linear units (GELUs) (2023), arXiv:1606.08415 [cs.LG]

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  48. [48]

    Zhang and R

    B. Zhang and R. Sennrich, Root mean square layer nor- malization, inAdvances in Neural Information Process- ing Systems,Vol.32, editedby H.Wallach, H.Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox, and R. Garnett (Curran Associates, Inc., 2019)

    2019

  49. [49]

    Xiong, Y

    R. Xiong, Y. Yang, D. He, K. Zheng, S. Zheng, C. Xing, H. Zhang, Y. Lan, L. Wang, and T. Liu, On layer normal- ization in the Transformer architecture, inProceedings 10 of the 37th International Conference on Machine Learn- ing, Proceedings of Machine Learning Research, Vol. 119, edited by H. D. III and A. Singh (PMLR, 2020) pp. 10524–10533

    2020

  50. [50]

    Cassella, H

    G. Cassella, H. Sutterud, S. Azadi, N. D. Drummond, D. Pfau, J. S. Spencer, and W. M. C. Foulkes, Discov- ering quantum phase transitions with fermionic neural networks, Phys. Rev. Lett.130, 036401 (2023)

    2023

  51. [51]

    Kato, On the eigenfunctions of many-particle systems in quantum mechanics, Communications on Pure and Applied Mathematics10, 151 (1957)

    T. Kato, On the eigenfunctions of many-particle systems in quantum mechanics, Communications on Pure and Applied Mathematics10, 151 (1957)

    1957

  52. [52]

    N. D. Drummond, M. D. Towler, and R. J. Needs, Jas- trow correlation factor for atoms, molecules, and solids, Phys. Rev. B70, 235119 (2004)

    2004

  53. [53]

    We fix this be- cause the wavefunction’s overall scale is pure gauge, so the linear kernel could always be rescaled to make the bias1

    The addition of1is structurally a bias. We fix this be- cause the wavefunction’s overall scale is pure gauge, so the linear kernel could always be rescaled to make the bias1

  54. [54]

    De Palo, S

    S. De Palo, S. Conti, and S. Moroni, Monte Carlo simu- lations of two-dimensional charged bosons, Phys. Rev. B 69, 035109 (2004)

    2004

  55. [55]

    B. K. Clark, M. Casula, and D. M. Ceperley, Hexatic and mesoscopic phases in a 2D quantum coulomb system, Phys. Rev. Lett.103, 055701 (2009)

    2009

  56. [56]

    Boninsegni, Bose one-component plasma in two di- mensions: A Monte Carlo study, Phys

    M. Boninsegni, Bose one-component plasma in two di- mensions: A Monte Carlo study, Phys. Rev. B113, 094508 (2026)

    2026

  57. [57]

    Geier, K

    M. Geier, K. Nazaryan, T. Zaklama, and L. Fu, Self- attention neural network for solving correlated electron problems in solids, Phys. Rev. B112, 045119 (2025)

    2025

  58. [58]

    L. H. Nosanow and G. L. Shaw, Hartree calculations for the ground state of solid He and other noble gas crystals, Phys. Rev.128, 546 (1962)

    1962

  59. [59]

    W. R. Magro and D. M. Ceperley, Ground-state prop- erties of the two-dimensional bose coulomb liquid, Phys. Rev. Lett.73, 826 (1994)

    1994

  60. [60]

    R. S. Sutton, The bitter lesson,http://www. incompleteideas.net/IncIdeas/BitterLesson.html (2019), accessed: 2026-06-09

    2019

  61. [61]

    Goodfellow, Y

    I. Goodfellow, Y. Bengio, and A. Courville,Deep Learn- ing(MIT Press, 2016)http://www.deeplearningbook. org

    2016

  62. [62]

    Westerhout, N

    T. Westerhout, N. Astrakhantsev, K. S. Tikhonov, M. I. Katsnelson, and A. A. Bagrov, Generalization properties of neural network approximations to frustrated magnet ground states, Nature Communications11, 1593 (2020)

    2020

  63. [63]

    Zhang, R

    H. Zhang, R. J. Webber, M. Lindsey, T. C. Berkelbach, and J. Weare, Understanding and eliminating spurious modes in variational Monte Carlo using collective vari- ables, Phys. Rev. Res.5, 023101 (2023)

    2023

  64. [64]

    Sorella, Green function Monte Carlo with stochastic reconfiguration, Phys

    S. Sorella, Green function Monte Carlo with stochastic reconfiguration, Phys. Rev. Lett.80, 4558 (1998)

    1998

  65. [65]

    Sorella, Generalized Lanczos algorithm for variational quantum Monte Carlo, Phys

    S. Sorella, Generalized Lanczos algorithm for variational quantum Monte Carlo, Phys. Rev. B64, 024512 (2001)

    2001

  66. [66]

    S.-i.Amari,Naturalgradientworksefficientlyinlearning, Neural Computation10, 251 (1998)

    1998

  67. [67]

    Chen and M

    A. Chen and M. Heyl, Empowering deep neural quantum states through efficient optimization, Nature Physics20, 1476 (2024)

    2024

  68. [68]

    Rende, L

    R. Rende, L. L. Viteritti, L. Bardone, F. Becca, and S. Goldt, A simple linear algebra identity to optimize large-scale neural network quantum states, Communica- tions Physics7, 260 (2024)

    2024

  69. [69]

    R. Li, H. Ye, D. Jiang, X. Wen, C. Wang, Z. Li, X. Li, D. He, J. Chen, W. Ren, and et al., A computational framework for neural network-based variational Monte Carlo with forward laplacian, Nature Machine Intelli- gence6, 209–219 (2024)

    2024

  70. [70]

    D. D. Dai and M. Soljačić, Essentially no energy bar- rier between independent fermionic neural quantum state minima (2026), arXiv:2601.06939 [cond-mat.dis-nn]

    work page arXiv 2026

  71. [71]

    Efron and R

    B. Efron and R. Tibshirani,An introduction to the boot- strap(Chapman et Hall/CRC, 1994)

    1994

  72. [72]

    K. T. Schütt, P.-J. Kindermans, H. E. Sauceda, S. Chmiela, A. Tkatchenko, and K.-R. Müller, Schnet: a continuous-filter convolutional neural network for model- ing quantum interactions, inProceedings of the 31st In- ternational Conference on Neural Information Processing Systems, NIPS’17 (Curran Associates Inc., Red Hook, NY, USA, 2017) p. 992–1002

    2017

  73. [73]

    Gerard, M

    L. Gerard, M. Scherbela, P. Marquetand, and P. Grohs, Gold-standard solutions to the Schrödinger equation us- ing deep learning: How much physics do we need?, inAd- vances in Neural Information Processing Systems, edited by A. H. Oh, A. Agarwal, D. Belgrave, and K. Cho (2022)

    2022

  74. [74]

    Pescia, J

    G. Pescia, J. Nys, J. Kim, A. Lovato, and G. Carleo, Message-passing neural quantum states for the homoge- neous electron gas, Phys. Rev. B110, 035108 (2024)

    2024

  75. [75]

    D. Luo, D. D. Dai, and L. Fu, Pairing-based graph neural network for simulating quantum materials, Phys. Rev. B 113, 165107 (2026)

    2026