Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Local Basis Transformation to Mitigate Negative Sign Problems

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that the negative sign problem in quantum Monte Carlo can be systematically mitigated by minimizing a locally computable L1 adaptive loss through local spin-basis rotations, with a proven upper bound for frustration-free…

desk verdict The L1 adaptive loss and the numerical work are worth attention, but Theorem 1 as stated is wrong by a factor |B| and Eq. (8) is unjustified; the paper needs a major revision before its central claim can be trusted. read the letter →

arxiv 2501.18069 v1 pith:IWNETXF6 submitted 2025-01-30 cond-mat.str-el cond-mat.stat-mech

Keisuke Murota , Synge Todo This is my paper · ORCID
classification cond-mat.str-elcond-mat.stat-mech
keywords negativesignproblemquantumMonteCarlobasisrotationfrustration-freeHamiltonianstoquasticL1adaptivelossunitarytransformationfrustratedspinsystems
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper's thesis is that the negative sign problem of quantum Monte Carlo is not an irreducible obstruction: it can be turned into an optimization objective over the representation basis. The authors introduce a metric (the negativity, and its temperature-independent form, the gap negativity) and prove that the usual fallback of simulating the absolutized stoquastic Hamiltonian is always the optimal virtual Hamiltonian, so only the basis remains to be chosen. Under the frustration-free assumption, they define the L1 adaptive loss, a sum of local bond terms that is independent of system size when the lattice is translationally invariant, and prove it is an upper bound on the gap negativity. Their numerical experiments show that optimizing local orthogonal or unitary rotations against this loss removes or substantially reduces the sign problem in random 1D and 2D frustration-free models, reproduces known sign-free dimer bases in the J0-J1-J2-J3 chain, bilinear-biquadratic chain, and Shastry-Sutherland model, and partially helps even in the strongly frustrated kagome Heisenberg model.

What carries the argument

The central object is the L1 adaptive loss, $L_{L1}(G,u) = |B|^{-1}\sum_{b\in B}[\lambda_+((u\otimes u)g_b(u^\dagger\otimes u^\dagger)) - \lambda(g_b)]$, with $G=\sum_b g_b=-H$ and $\lambda_+$ the largest eigenvalue after taking absolute values elementwise. Rewriting the sum as a maximum over normalized non-negative states $|\psi\rangle$ of $\sum_b \sum_{i,j} \psi_i\psi_j^* |((u\otimes u)g_b(u^\dagger\otimes u^\dagger))_{ij}|$ turns basis optimization into a min-max problem. Perron-Frobenius lets the maximizing state be taken non-negative; the frustration-free property (one common eigenstate for all $g_b$) is what makes the inequality $L_{L1}\ge \lambda_+(G)-\lambda(G)$ hold, so the loss is a computable, system-size-independent surrogate for the gap negativity. Riemannian gradient descent with Adam optimizes $u$ under the constraint $U=\otimes_i u_i$.

What would settle it

Take a small translation-invariant frustration-free chain (e.g., an L=6 random parent Hamiltonian of the kind used in Fig. 2), minimize the L1 adaptive loss over all allowed local unitaries, and then compute the true gap negativity $\lambda_+(G)-\lambda(G)$ by exact diagonalization of the rotated Hamiltonian; finding any case with $\lambda_+(G)-\lambda(G) > L_{L1}(G,u)$ would refute Theorem 1. For the method's usefulness outside the frustration-free regime, run the kagome anisotropic Heisenberg model at $J_z=1, h_x=0$ with $\beta=0.5$ and compare the average sign before and after loss minimization: if the sign worsens while the loss drops, the surrogate has detached from the true sign severity.

Watch

Extended reading notes

Core claim

The central claim is that gap negativity $\eta' = \lambda_+(G)-\lambda(G)$, the exponential rate at which the average sign decays, can be bounded from above by the L1 adaptive loss $L_{L1}(G,u)$ whenever $G=\sum_b g_b$ is frustration-free. The theorem $L_{L1}(G,u)\ge \lambda_+(G)-\lambda(G)$ means that a quantity computable from local, system-size-independent data controls the global severity of the sign problem, so reducing the loss through local basis rotations is a safe strategy. The numerical results support that this is more than an inequality: in frustration-free benchmarks the optimized rotations either reach $\eta'=0$ (complete sign removal) or approach it, and in non-frustration-free cases such as the kagome Heisenberg model the method still lowers the sign problem in some parameter regimes. The paper also finds that allowing unitary rather than only orthogonal local rotations can restore symmetries and further reduce the loss, even though the original Hamiltonian is real symmetric.

Load-bearing premise

The load-bearing premise is that the Hamiltonian is frustration-free: a single state is a simultaneous eigenstate of every bond term, and this same non-negative state realizes the Perron-Frobenius maxima for all bonds at once; if that shared-state property fails, the L1 adaptive loss is only a heuristic and the upper-bound guarantee collapses.

Editorial extensions

If this is right

  • For any frustration-free Hamiltonian, the L1 adaptive loss is a safe objective: because it upper-bounds the gap negativity, lowering it cannot make the true sign problem worse than the loss itself.
  • The optimal-virtual-Hamiltonian proposition reduces sign-problem mitigation to basis selection alone, so future methods need not spend effort designing reweighting Hamiltonians.
  • The loss is system-size independent under translational invariance, which means the same optimized local rotation can be applied to arbitrarily large lattices without re-solving the optimization.
  • Numerically, the method reproduces analytically constructed sign-free bases for the J0-J1-J2-J3 chain, the bilinear-biquadratic chain, and the Shastry-Sutherland model, showing a common mechanism behind previously ad hoc transformations.
  • Allowing unitary local rotations can go beyond orthogonal ones, restoring lattice symmetries and further reducing the sign problem in models like the kagome Heisenberg antiferromagnet.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper's claims, the upper-bound theorem suggests frustration-free-like structure is the regime where local losses can be reliable; for strongly frustrated systems, a hierarchical or adaptive cell partition that shrinks the cell-boundary terms noted in Remark 1 might be a natural extension.
  • Editorial extension: the min-max form of the loss resembles a local ground-state problem, so one testable next step is to vary the cell size and measure whether the optimized gap negativity improves monotonically, cleanly separating the approximation error of the loss from the optimization error.
  • Editorial extension: because unitary transformations help by breaking real-symmetry constraints, a concrete test would be to compare optimized complex phases against loop frustration patterns in the kagome model, to see whether they cancel sign-inducing amplitudes rather than just reducing their magnitude.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proposes a systematic local-basis-transformation approach to mitigating the negative sign problem in quantum Monte Carlo simulations of frustrated spin systems. It defines a negativity measure, proves that the stoquastic Hamiltonian G+ is the optimal virtual Hamiltonian in a reweighting scheme (Proposition 1), and introduces an L1 adaptive loss function for local basis optimization. The central theoretical claim is Theorem 1, which states that this loss upper-bounds the gap negativity λ_+(G)-λ(G) for frustration-free systems. The numerical sections apply the method to random 1D and 2D frustration-free parent Hamiltonians, the J0-J1-J2-J3 model, the bilinear-biquadratic chain, the Shastry-Sutherland model, and the kagome Heisenberg model, and compare orthogonal versus unitary local transformations. The paper reports substantial reductions of the gap negativity and recovery of several known sign-free basis choices.

Significance. If Theorem 1 were correct, the paper would be a significant contribution: it would provide a local, system-size-independent, computationally tractable loss function whose minimization provably reduces the sign problem, including in 2D systems. The numerical work is extensive and, as far as the manuscript shows, honestly executed: the gap negativity is evaluated by independent exact diagonalization, no fitted parameters are introduced, and the tests reproduce analytically known sign-free regions. Proposition 1 is a useful and correct formal observation. However, the main formal guarantee is not correct as stated. Theorem 1 drops a factor of |B|, Eq. (13) has the wrong sign/direction, and the equality in Eq. (8) between the sum of bond-wise Perron eigenvalues and the adaptive max-over-|ψ⟩ loss is unjustified. These are not presentation issues: they affect the paper's central claim that minimizing the proposed loss certifiably reduces the extensive gap negativity. The numerical demonstrations are promising, but they do not repair the broken certificate.

major comments (3)
  1. [IV.C, Theorem 1 and Eqs. (11)-(13)] The stated inequality L_L1(G,u) ≥ λ_+(G)-λ(G) does not follow from the proof. From Eq. (12) and frustration-freeness one obtains λ_+(G)-λ(G) ≤ Σ_b [λ_+(g_b)-λ(g_b)] = |B| L_L1(G,u), i.e., L_L1(G,u) ≥ (λ_+(G)-λ(G))/|B|. The proof therefore establishes only a factor-|B|-weaker bound. Moreover, Eq. (13) has the wrong direction and sign: because G+ is entrywise larger than G, the Weyl monotonicity theorem gives λ_+(G) ≥ λ(G), so λ(G)-λ_+(G) ≤ 0, while the proof requires λ(G)-λ_+(G) ≥ 0. This is load-bearing because L_L1 is an intensive per-bond average while λ_+-λ is extensive; the missing factor depends on system size. The central claim that minimizing L_L1 certifiably reduces the gap negativity is therefore unsupported.
  2. [IV.C, Eq. (8)] The equality (1/|B|) Σ_b λ_+((u⊗u)g_b(u†⊗u†)) = max_ψ (1/|B|) Σ_b Σ_ij ψ_i ψ_j^* |((u⊗u)g_b(u†⊗u†))_ij| would require a single normalized state |ψ⟩ that simultaneously saturates the Perron-Frobenius variational bound for every bond. Frustration-freeness (Definition 1) supplies a common eigenstate of the original operators g_b, but not of their absolute-value matrices (g_b)_+. The Perron vectors of the different (g_b)_+ will generally differ from bond to bond. Consequently, the quantity actually optimized by the adaptive max-over-|ψ⟩ procedure is not provably equal to the L_L1 used in Theorem 1; it is, in fact, pointwise no larger than L_L1, so it can be smaller than any bound derived for L_L1.
  3. [V.A, Fig. 3] The numerical data in Fig. 3 are consistent with the factor-|B| bound and not with Theorem 1 as stated. The authors draw the line y=6x, explicitly noting |B|=6, and state that most points lie below this line. This is exactly the relation Δ_improvement ≤ |B| L_L1_improvement, not Δ_improvement ≤ L_L1_improvement. The text's claim that this linear relationship is a consequence of Theorem 1 is therefore misleading: the slope 6 is precisely the factor missing from the theorem. This is not a minor plotting detail but a direct numerical confirmation that the stated upper bound is false.
minor comments (4)
  1. [IV.C, Remark 1] The cross-reference to Eq. (10) should be to Eq. (9), and "servers" should be "serves" in the sentence "it servers as a local Hamiltonian."
  2. [V.B.1] The sentence beginning "The reason for the apparent persistence of the negative sign in this region is that, although log(η)/β reaches the minimum value after optimization, the degeneracy of the ground state increases compared to the original Hamiltonian G" is duplicated verbatim.
  3. [V.B.3] The model name is misspelled as "Shustry-Sutherland" in the text; it should be "Shastry-Sutherland."
  4. [V.C] The phrase "in certain parameter region Fig. 11" should read "in the parameter region shown in Fig. 11" (or similar) for clarity.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the L1 adaptive loss is minimized against an independently evaluated global spectral quantity, and the suspected Theorem 1 issues are derivation gaps rather than circular reductions.

full rationale

I find no circular derivation chain in this paper. The L1 adaptive loss (Definition 2) is constructed from local spectral quantities, while the target gap negativity η' = λ_+(G) − λ(G) is an independent, system-wide spectral quantity that is evaluated by exact diagonalization in the numerical sections. The optimization procedure minimizes the loss over local basis rotations, and the reported improvements are measured against the independently computed gap negativity, not against the loss itself. No fitted parameter is renamed as a prediction: the loss is not calibrated to the gap-negativity data used for benchmarking. The authors' self-citations (e.g., Refs. [5], [38], [39] concerning QMC update schemes) are algorithmic and not load-bearing for the central claim, and Lemma 1 is attributed to the external textbook Ref. [32]. The proof of Theorem 1 does contain a likely factor-of-|B| gap: the stated inequality L_L1 ≥ λ_+(G) − λ(G) is stronger than what the subadditivity argument yields, and Eq. (8)'s replacement of the sum of local Perron roots by a single max-over-ψ expression is not guaranteed in general. These are mathematical derivation gaps or correctness errors, not circularity: the theorem does not assume its conclusion, and the numerical demonstrations reproduce known sign-free regimes as external checks. Accordingly, the appropriate circularity score is low.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the frustration-free assumption, the Perron-Frobenius theorem, the restriction to local tensor-product unitaries, and an unstated entrywise inequality. No free physical parameters are fitted to data; the only tunables are optimization hyperparameters and the choice of cell size.

assumptions (4)
  • domain assumption The Hamiltonian is frustration-free (Definition 1), so that λ(G) = Σ_b λ(g_b).
    Used in the proof of Theorem 1 to relate the global ground energy to the sum of local ground energies; not valid for generic frustrated systems.
  • standard math Perron-Frobenius theorem: the largest eigenvector of a stoquastic matrix can be chosen non-negative.
    Invoked in Section IV.C to justify the max over a non-negative state |ψ⟩ in the loss function.
  • domain assumption The local basis transformation is restricted to a product of identical local unitaries, U = ⊗_i u_i.
    Equation (7). Needed for computational tractability but limits the reachable basis set.
  • standard math Inequality λ_+(G) ≤ Σ_b λ_+(g_b) (Eq. 12) holds.
    Implied by entrywise domination G_+ ≤ Σ_b (g_b)_+; used without proof in the proof of Theorem 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Local Basis Transformation to Mitigate Negative Sign Problems." pith.science (2026). https://pith.science/paper/IWNETXF6

@misc{pith2026250118069,
  author       = {Pith},
  title        = {Pith review of: Local Basis Transformation to Mitigate Negative Sign Problems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IWNETXF6}},
  note         = {Machine review of arXiv:2501.18069}
}
read the original abstract

Quantum Monte Carlo (QMC) methods for the frustrated quantum spin systems occasionally suffer from the negative sign problem, which makes simulations exponentially harder for larger systems at lower temperatures and severely limits QMC's application across a wide range of spin systems. This problem is known to depend on the choice of representation basis. We propose a systematic approach for mitigating the sign problem independent of the given Hamiltonian or lattice structure. We first introduce the concept of negativity to characterize the severity of the negative sign problem. We then demonstrate the existence of a locally defined quantity, the L1 adaptive loss function, which effectively approximates negativity, especially in frustration-free systems. Using the proposed loss function, we demonstrate that optimizing the representation basis can mitigate the negative sign. This is evidenced by several frustration-free models and other important quantum spin systems. Furthermore, we compare the effectiveness of unitary transformations against the standard orthogonal transformation and reveal that unitary transformations can effectively mitigate the sign problem in certain cases.

Figures

Figures reproduced from arXiv: 2501.18069 by the authors.

Figure 1
Figure 1. FIG. 1. Illustration of the unitary transformation and the [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Histogram of gap negativity [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Scatter plot of improvements in gap negativity [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Histogram of gap negativity [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Lattice structure and cell for basis transformation [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Gap negativity [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 6
Figure 6. Figure 6: , the negative sign is mitigated for J3 < J2, while the region J3 > 2 ∩ J3 > J2 still appears to be af￾fected by the negative sign. The reason for the apparent persistence of the negative sign in this region is that, although log(η)/β reaches the minimum value after op…
Figure 8
Figure 8. Figure 8: FIG. 8. Gap negativity [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Shastry-Sutherland model defined on the 2D square [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 11
Figure 11. Figure 11: FIG. 11. Gap negativity [PITH_FULL_IMAGE:figures/full_fig_p009_11.png]
Figure 10
Figure 10. Figure 10: FIG. 10. Gap negativity [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: This issue, which is explored in more detail in the next section, arises because we permit only orthogo￾nal transformations as the local basis transformations [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12. Lattice structure of the kagome Heisenberg model [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]
Figure 13
Figure 13. Figure 13: FIG. 13. Comparison of L1 adaptive loss function with [PITH_FULL_IMAGE:figures/full_fig_p011_13.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Sign-optimized Quantum Monte Carlo

    cond-mat.str-el 2026-07 accept novelty 6.0 of 10

    Minimizing the phase of off-diagonal bond-Hamiltonian elements via local unitary rotations yields bases with better QMC average sign than computational or cluster eigenbases on frustrated Heisenberg models.

Reference graph

Works this paper leans on

61 extracted references · 61 canonical work pages · cited by 1 Pith paper

  1. [1]

    We choose the cells for local orthogonal ba- sis transformation as shown in Fig

    J0-J1-J2-J3 Model First, we apply our method to the J0-J1-J2-J3 model, which is a spin-1/2 Heisenberg chain with nearest- neighbor antiferromagnetic interactions J2 and J3 and next-nearest-neighbor antiferromagnetic interactions J0 and J1. We choose the cells for local orthogonal ba- sis transformation as shown in Fig. 5. In the simula- tion, J0 and J1 ar...

    work page

  2. [2]

    Bilinear-Biquadratic Chain Next, we consider the S = 1 bilinear-biquadratic (BLBQ) chain model. The Hamiltonian encapsulates a range of significant models, with the parameter α tun- ing the system through various quantum states from the nearest-neighbor antiferromagnetic Heisenberg chain (α = 0) to the AKLT model ( α = 1/3) and even to the intriguing SU(3...

    work page

  3. [3]

    The lattice structure and the cell for the local basis transformation are shown in Fig

    Shastry-Sutherland Model As a final example of the frustration-free models, we examine the Shastry-Sutherland model, a 2D frus- trated spin model with known negative-sign-free local basis transformation [19, 24]. The lattice structure and the cell for the local basis transformation are shown in Fig. 9. This model exhibits a wide variety of quan- tum phase...

    work page

  4. [4]

    Cluster algorithm for vertex models

    Hans Gerd Evertz, Gideon Lana, and Mihai Marcu. Cluster algorithm for vertex models. Phys. Rev. Lett., 70(7):875–879, February 1993

    work page 1993

  5. [5]

    Suzuki, editor

    M. Suzuki, editor. Quantum Monte Carlo Methods in Condensed Matter Physics. World Scientific, Singapore, 1994

    work page 1994

  6. [6]

    #$%"&−𝜂'#(

    The blue histogram represents the gap negativity of the original Hamiltonian, and the red histogram represents the gap negativity of the optimized Hamiltonian. Optimization is done by minimizing the L1 adaptive loss function for cell type (a) in Fig. 1. Evaluation of the gap negativity is done by exact diagonalization of the system Hamiltonian and its sto...

    work page

  7. [7]

    Strongly Correlated Systems: Numerical Methods, volume 176 of Springer Series in Solid-State Sciences

    Adolfo Avella and Ferdinando Mancini, editors. Strongly Correlated Systems: Numerical Methods, volume 176 of Springer Series in Solid-State Sciences. Springer, Berlin, Heidelberg, 2013

    work page 2013

  8. [8]

    Gubernatis, N

    J. Gubernatis, N. Kawashima, and P. Werner. Quan- 12 tum Monte Carlo Methods: Algorithms for Lattice Mod- els. Cambridge University Press, 2016

    work page 2016

Show all 61 references
  1. [9]

    Loop Algorithm

    Synge Todo. Loop Algorithm. In Adolfo Avella and Fer- dinando Mancini, editors, Strongly Correlated Systems: Numerical Methods, pages 153–184. Springer, Berlin, Hei- delberg, 2013

    2013

  2. [10]

    N. V. Prokof’ev, B. V. Svistunov, and I. S. Tupitsyn. Exact, complete, and universal continuous-time world- line Monte Carlo approach to the statistics of discrete quantum systems. Soviet Physics JETP, 87:310, 1998

    1998

  3. [11]

    O. F. Syljuasen and A. W. Sandvik. Quantum Monte Carlo with directed loops. Phys. Rev. E, 66:046701, 2002

    2002

  4. [12]

    Quantum Monte Carlo simulation method for spin systems

    A W Sandvik and J Kurkij¨ arvi. Quantum Monte Carlo simulation method for spin systems. Phys. Rev. B, 43(7):5950–5961, March 1991

    1991

  5. [13]

    Reweighting method for quantum Monte Carlo simulations with the negative-sign problem

    Tota Nakamura, Naomichi Hatano, and Hidetoshi Nishi- mori. Reweighting method for quantum Monte Carlo simulations with the negative-sign problem. J. Phys. Soc. Jpn., 61(10):3494–3502, October 1992

    1992

  6. [14]

    Data analysis for quantum Monte Carlo simulations with the negative-sign problem

    Naomichi Hatano. Data analysis for quantum Monte Carlo simulations with the negative-sign problem. J. Phys. Soc. Jpn., 63(5):1691–1697, May 1994

    1994

  7. [15]

    Vanishing of the negative-sign problem of quantum monte carlo simulations in one-dimensional frustrated spin systems

    Tota Nakamura. Vanishing of the negative-sign problem of quantum monte carlo simulations in one-dimensional frustrated spin systems. Phys. Rev. B, 57(6):R3197– R3200, February 1998

    1998

  8. [16]

    Easing the Monte Carlo sign problem

    Dominik Hangleiter, Ingo Roth, Daniel Nagaj, and Jens Eisert. Easing the Monte Carlo sign problem. Sci. Adv., 6(33):eabb8341, 2020

    2020

  9. [17]

    Wallman, and Stephen D

    Hakop Pashayan, Joel J. Wallman, and Stephen D. Bartlett. Estimating outcome probabilities of quan- tum circuits using quasiprobabilities. Phys. Rev. Lett., 115(7):070501, 2015

    2015

  10. [18]

    Representation basis in quantum Monte Carlo calculations and the negative-sign problem

    Naomichi Hatano and Masuo Suzuki. Representation basis in quantum Monte Carlo calculations and the negative-sign problem. Phys. Lett. A, 163(4):246–249, 1992

    1992

  11. [19]

    Ryan Levy and Bryan K. Clark. Mitigating the sign problem through basis rotations. Phys. Rev. Lett. , 126(21):216401, 2021

    2021

  12. [20]

    Haldane and dimer gap in general double spin-chain models

    Tota Nakamura, Satoshi Takada, Kiyomi Okamoto, and Naoki Kurosawa. Haldane and dimer gap in general double spin-chain models. J. Phys.: Condens. Matter, 9(30):6401, jul 1997

    1997

  13. [21]

    Vanishing of the negative-sign problem of quantum Monte carlo simulations in one-dimensional frustrated spin systems, July 1997

    Tota Nakamura. Vanishing of the negative-sign problem of quantum Monte carlo simulations in one-dimensional frustrated spin systems, July 1997

    1997

  14. [22]

    Normand, Fr´ ed´ eric Mila, and Andreas Honecker

    Stefan Wessel, B. Normand, Fr´ ed´ eric Mila, and Andreas Honecker. Efficient quantum Monte Carlo simulations of highly frustrated magnets: The frustrated spin-1/2 ladder. SciPost Phys., 3:005, 2017

    2017

  15. [23]

    Nor- mand, Fr´ ed´ eric Mila, Philippe Corboz, and Andreas Honecker

    Stefan Wessel, Ido Niesen, Jonas Stapmanns, B. Nor- mand, Fr´ ed´ eric Mila, Philippe Corboz, and Andreas Honecker. Thermodynamic properties of the Shastry- Sutherland model from quantum Monte Carlo simula- tions. Phys. Rev. B, 98:174432, 2018

    2018

  16. [24]

    Symmetry- protected topological order and negative-sign problem for SO( N ) bilinear-biquadratic chains

    Kouichi Okunishi and Kenji Harada. Symmetry- protected topological order and negative-sign problem for SO( N ) bilinear-biquadratic chains. Phys. Rev. B, 89:134422, 2014

    2014

  17. [25]

    On the computational complexity of curing non-stoquastic Hamiltonians

    Milad Marvian, Daniel A Lidar, and Itay Hen. On the computational complexity of curing non-stoquastic Hamiltonians. Nat. Commun., 10(1):1571, April 2019

    2019

  18. [26]

    Hardness and ease of curing the sign problem for two-local qubit Hamiltonians

    Joel Klassen, Milad Marvian, Stephen Piddock, Marios Ioannou, Itay Hen, and Barbara M Terhal. Hardness and ease of curing the sign problem for two-local qubit Hamiltonians. SIAM J. Comput., 49(6):1332–1362, Jan- uary 2020

    2020

  19. [27]

    Sign- problem-free Monte Carlo simulation of certain frus- trated quantum magnets

    Fabien Alet, Kedar Damle, and Sumiran Pujari. Sign- problem-free Monte Carlo simulation of certain frus- trated quantum magnets. Phys. Rev. Lett., 117:197203, 2016

    2016

  20. [28]

    Exact dimer ground state of the two dimensional heisenberg spin system SrCu2(BO3)2

    Shin Miyahara and Kazuo Ueda. Exact dimer ground state of the two dimensional heisenberg spin system SrCu2(BO3)2. Phys. Rev. Lett., 82(18):3701–3704, 1999

    1999

  21. [29]

    Multiple magnetization plateaus induced by farther neighbor interaction in an S = 1 two-leg Heisenberg spin ladder, February 2021

    Hidehiko Kohshiro, Ryui Kaneko, Satoshi Morita, Hosho Katsura, and Naoki Kawashima. Multiple magnetization plateaus induced by farther neighbor interaction in an S = 1 two-leg Heisenberg spin ladder, February 2021

    2021

  22. [30]

    Honecker, S

    A. Honecker, S. Wessel, R. Kerkdyk, T. Pruschke, F. Mila, and B. Normand. Thermodynamic properties of highly frustrated quantum spin ladders: Influence of many-particle bound states. Phys. Rev. B, 93(5):054408, February 2016

    2016

  23. [31]

    The ANNNI model — theoretical analysis and experimental application

    Walter Selke. The ANNNI model — theoretical analysis and experimental application. Phys. Rep., 170(4):213– 264, November 1988

    1988

  24. [32]

    Calculation of spin correlations in two-dimensional Ising systems from one-dimensional kinetic models

    I Peschel and V J Emery. Calculation of spin correlations in two-dimensional Ising systems from one-dimensional kinetic models. Z. Phys. B: Condens. Matter, 43(3):241– 249, September 1981

    1981

  25. [33]

    Yu Kitaev

    A. Yu Kitaev. Fault-tolerant quantum computation by anyons, 2003

    2003

  26. [34]

    Superconductivity and the quantum hard-core dimer gas

    D S Rokhsar and S A Kivelson. Superconductivity and the quantum hard-core dimer gas. Phys. Rev. Lett., 61(20):2376–2379, November 1988

    1988

  27. [35]

    Valence bond ground states in isotropic quantum antiferromagnets

    Ian Affleck, Tom Kennedy, and Hal Lieb, Elliott Gand Tasaki. Valence bond ground states in isotropic quantum antiferromagnets. Commun. Math. Phys. , 115(3):477–528, September 1988

    1988

  28. [36]

    Physics and Mathematics of Quantum Many- Body Systems

    Hal Tasaki. Physics and Mathematics of Quantum Many- Body Systems. Springer, 2020

    2020

  29. [37]

    Zur theorie der matrices

    Oskar Perron. Zur theorie der matrices. Math. Ann., 64(2):248–263, June 1907

    1907

  30. [38]

    An SU(2)- symmetric semidefinite programming hierarchy for quan- tum max cut, 2023

    Jun Takahashi, Chaithanya Rayudu, Cunlu Zhou, Rob- bie King, Kevin Thompson, and Ojas Parekh. An SU(2)- symmetric semidefinite programming hierarchy for quan- tum max cut, 2023

    2023

  31. [39]

    Monte Carlo Sta- tistical Methods

    Christian Robert and George Casella. Monte Carlo Sta- tistical Methods. Springer texts in statistics. Springer, New York, NY, November 2010

    2010

  32. [40]

    A guide to Monte Carlo simulations in statistical physics

    David P Landau and Kurt Binder. A guide to Monte Carlo simulations in statistical physics. Cambridge Uni- versity Press, Cambridge, England, 4 edition, November 2014

    2014

  33. [41]

    Monte Carlo Methods in Statistical Physics

    M E J Newman and G T Barkema. Monte Carlo Methods in Statistical Physics. Oxford University Press, 02 1999

    1999

  34. [42]

    Geometric Allocation Approach for the Transition Kernel of a Markov Chain, pages 213–222

    Hidemaro Suwa and Synge Todo. Geometric Allocation Approach for the Transition Kernel of a Markov Chain, pages 213–222. De Gruyter, Berlin, Boston, 2013

    2013

  35. [43]

    Geometrically Constructed Markov Chain Monte Carlo Study of Quantum Spin-Phonon Complex Systems

    Hidemaro Suwa. Geometrically Constructed Markov Chain Monte Carlo Study of Quantum Spin-Phonon Complex Systems. Springer Japan, 2012

    2012

  36. [44]

    Steepest descent algorithms for optimization under uni- 13 tary matrix constraint

    Traian E Abrudan, Jan Eriksson, and Visa Koivunen. Steepest descent algorithms for optimization under uni- 13 tary matrix constraint. IEEE Trans. Signal Process., 56(3):1134–1147, April 2008

    2008

  37. [45]

    Con- jugate gradient algorithm for optimization under unitary matrix constraint

    Traian Abrudan, Jan Eriksson, and Visa Koivunen. Con- jugate gradient algorithm for optimization under unitary matrix constraint. Signal Processing, 89(9):1704–1714, September 2009

    2009

  38. [46]

    Trivializations for gradient- based optimization on manifolds

    Mario Lezcano-Casado. Trivializations for gradient- based optimization on manifolds. In Advances in Neural Information Processing Systems (NeurIPS), pages 9154– 9164, 2019

    2019

  39. [47]

    Cheap orthogonal constraints in neural networks: A simple parametrization of the orthogonal and unitary group

    Mario Lezcano-Casado and David Mart ´ ınez-Rubio. Cheap orthogonal constraints in neural networks: A simple parametrization of the orthogonal and unitary group. In International Conference on Machine Learning (ICML), pages 3794–3803, 2019

    2019

  40. [48]

    Adam: A method for stochastic optimization, December 2014

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization, December 2014

    2014

  41. [49]

    Perez-Garcia, F

    D. Perez-Garcia, F. Verstraete, M. M. Wolf, and J. I. Cirac. Matrix product state representations. Quantum Inf. Comput., 7:401–430, 2007

    2007

  42. [50]

    Frustra- tion free gapless Hamiltonians for matrix product states

    Carlos Fern´ andez-Gonz´ alez, Norbert Schuch, Michael M Wolf, J Ignacio Cirac, and David P´ erez-Garc ´ ıa. Frustra- tion free gapless Hamiltonians for matrix product states. Commun. Math. Phys., 333:299–333, 2015

    2015

  43. [51]

    Matrix product ground states for one-dimensional spin-1 quan- tum antiferromagnets

    A Kl¨ umper, A Schadschneider, and J Zittartz. Matrix product ground states for one-dimensional spin-1 quan- tum antiferromagnets. EPL, 24(4):293–297, November 1993

    1993

  44. [52]

    A practical introduction to tensor net- works: Matrix product states and projected entangled pair states

    Rom´ an Or´ us. A practical introduction to tensor net- works: Matrix product states and projected entangled pair states. Ann. Phys., 349:117–158, October 2014

    2014

  45. [53]

    Ignacio Cirac, and Michael M

    David Perez-Garcia, Frank Verstraete, J. Ignacio Cirac, and Michael M. Wolf. PEPS as unique ground states of local Hamiltonians. Quantum Inf. Comput., 8:0650–0663, 2008

    2008

  46. [54]

    PEPS as ground states: Degeneracy and topology

    Norbert Schuch, Ignacio Cirac, and David Perez-Garcia. PEPS as ground states: Degeneracy and topology. Ann. Phys., 325:2153–2192, 2010

    2010

  47. [55]

    Model for a multicomponent quantum system

    Bill Sutherland. Model for a multicomponent quantum system. Phys. Rev. B, 12:3795, 1975

    1975

  48. [56]

    Quantum phase transitions in the Shastry-Sutherland model for SrCu 2(BO3)2

    A Koga and N Kawakami. Quantum phase transitions in the Shastry-Sutherland model for SrCu 2(BO3)2. Phys. Rev. Lett., 84(19):4461–4464, May 2000

    2000

  49. [57]

    Quan- tum criticality and spin liquid phase in the Shastry- Sutherland model, 2022

    Jianwei Yang, Anders W Sandvik, and Ling Wang. Quan- tum criticality and spin liquid phase in the Shastry- Sutherland model, 2022

    2022

  50. [58]

    Properties of an algebraic spin liquid on the kagome lattice

    Michael Hermele, Ying Ran, Patrick A Lee, and Xiao- Gang Wen. Properties of an algebraic spin liquid on the kagome lattice. Phys. Rev. B, 77(22):224413, June 2008

    2008

  51. [59]

    Physics of low-energy singlet states of the kagome lattice quantum Heisenberg antifer- romagnet

    P Nikolic and T Senthil. Physics of low-energy singlet states of the kagome lattice quantum Heisenberg antifer- romagnet. Phys. Rev. B, 68(21):214415, December 2003

    2003

  52. [60]

    Magnetization process of kagome-lattice heisenberg antiferromagnet

    Hiroki Nakano and Toru Sakai. Magnetization process of kagome-lattice heisenberg antiferromagnet. J. Phys. Soc. Jpn., 79(5):053707, 2010

    2010

  53. [61]

    Reduction of the sign problem near T = 0 in quantum Monte Carlo simulations

    Jonathan D’Emidio, Stefan Wessel, and Fr´ ed´ eric Mila. Reduction of the sign problem near T = 0 in quantum Monte Carlo simulations. Phys. Rev. B, 102(6):064420, August 2020

    2020

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.