pith. sign in

arxiv: 2505.21255 · v3 · submitted 2025-05-27 · 🪐 quant-ph

Quantum Markov chain Monte Carlo method with programmable quantum simulators

Mauro D'Arcangelo , Younes Javanmard , Natalie Pearson This is my paper

Pith reviewed 2026-05-19 12:57 UTC · model grok-4.3

classification 🪐 quant-ph
keywords quantum Markov chain Monte Carlomany-body localizationquantum simulatorscombinatorial optimizationFloquet dynamicsIsing chainthermal sampling
0
0 comments X p. Extension

The pith

A quantum Markov chain Monte Carlo method uses the many-body localized phase to sample thermal distributions on programmable simulators.

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

The paper presents a quantum Markov chain Monte Carlo algorithm that runs on many-body systems by exploiting the many-body localized phase. This phase supplies the ergodicity needed for proper sampling while the transition to localization provides a way to adjust the rate at which proposed states are accepted. The approach applies directly to combinatorial optimization problems and requires only the ability to simulate driven one-dimensional Ising chains. A reader would care because it opens a route to generating thermal samples for Hamiltonians that existing quantum hardware cannot implement directly.

Core claim

The algorithm relies on the many-body localized phase to meet the conditions for ergodicity and to draw samples from target distributions of quantum states, with the thermalized-to-localized transition used to control the acceptance probability of the Markov chain; this construction works for quadratic and higher-order optimization problems and runs on any processor that can implement Floquet dynamics of a nearest-neighbor Ising chain.

What carries the argument

The many-body localized (MBL) phase, which supplies ergodicity for the Markov chain and permits tuning of acceptance rates through the localization transition.

If this is right

  • The method solves combinatorial optimization problems of quadratic order and higher.
  • Any quantum processor able to simulate Floquet evolution of a one-dimensional Ising chain with nearest-neighbor couplings can run the algorithm.
  • It enables sampling from thermal distributions of Hamiltonians that cannot be realized natively on the available hardware.

Where Pith is reading between the lines

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

  • Hardware platforms limited to Ising-type interactions could still address sampling tasks for a wider class of problems than their native Hamiltonians allow.
  • The same localization-based control of acceptance might be adapted to other quantum Monte Carlo variants that currently suffer from poor mixing.
  • Benchmarking the method on small optimization instances would reveal how the tunable acceptance rate affects convergence speed relative to classical MCMC.

Load-bearing premise

The many-body localized phase supplies the ergodicity and acceptance-rate control required for the Markov chain to converge to the desired thermal distribution.

What would settle it

Numerical or experimental observation that the long-time distribution generated by the algorithm deviates from the target Boltzmann distribution when the driving parameters place the system outside the MBL regime.

Figures

Figures reproduced from arXiv: 2505.21255 by Mauro D'Arcangelo, Natalie Pearson, Younes Javanmard.

Figure 1
Figure 1. Figure 1: FIG. 1: The figure illustrates the schematics of the [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Numerical simulation of the thermalization of a [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Probability of observing any of the optimal [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Probability of observing the optimal solution to [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: Convergence of products of MBL unitaries to [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Thermalization of a system of 9 qubits under [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Effect of driving amplitude on the [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
read the original abstract

In this work, we present a quantum Markov chain algorithm for many-body systems that utilizes a special phase of matter known as the Many-Body Localized (MBL) phase. We show how the properties of the MBL phase enable one to address the conditions for ergodicity and sampling from distributions of quantum states. We demonstrate how to exploit the thermalized-to-localized transition to tune the acceptance rate of the Markov chain, and apply the algorithm to solve a range of combinatorial optimization problems of quadratic order and higher. The algorithm can be implemented on any quantum processing unit capable of simulating the Floquet dynamics of a one-dimensional Ising chain with nearest-neighbor interactions, providing a practical way of sampling from thermal distributions of Hamiltonians that cannot be natively implemented on the quantum hardware.

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

2 major / 2 minor

Summary. The manuscript proposes a quantum Markov chain Monte Carlo (QMCMC) algorithm that exploits the many-body localized (MBL) phase of a one-dimensional Ising chain under Floquet driving. It claims that MBL properties address ergodicity requirements for sampling, that the thermalized-to-localized transition can be used to tune the Metropolis acceptance rate, and that the resulting sampler can be realized on any programmable quantum simulator capable of nearest-neighbor Ising Floquet evolution, thereby enabling thermal sampling of Hamiltonians that cannot be natively implemented on the hardware. The method is applied to quadratic and higher-order combinatorial optimization problems.

Significance. If the irreducibility of the induced Markov chain and convergence to the target Boltzmann distribution can be established, the approach would offer a concrete route to leverage existing 1D Ising simulators for sampling tasks beyond the native Hamiltonian, which is a practically relevant capability for quantum optimization and statistical mechanics. The use of an MBL transition to control acceptance rates is a distinctive idea that, if substantiated, could stimulate further work on phase-of-matter-assisted quantum algorithms.

major comments (2)
  1. [Abstract / central construction] Abstract and central construction (method description): The manuscript asserts that MBL properties 'enable one to address the conditions for ergodicity.' However, MBL eigenstates are localized with exponentially decaying correlations and the dynamics are non-ergodic by definition. No explicit argument or derivation is supplied showing that the Floquet evolution in the MBL regime generates a transition kernel that is irreducible over the full configuration space of an arbitrary target Hamiltonian. Without such a demonstration (e.g., a lower bound on the spectral gap or a connectivity argument), the stationary distribution reached by the sampler may be supported only on a restricted subset of states rather than the desired thermal distribution.
  2. [Method section] Method section (description of the acceptance-rate tuning): The claim that the thermalized-to-localized transition can be exploited to tune the acceptance rate of the Markov chain is load-bearing for practical performance. The manuscript must specify how the localization length or driving parameters are chosen so that the effective proposal distribution remains sufficiently connected while still satisfying detailed balance with respect to the target Boltzmann weight; otherwise the tuning may inadvertently suppress transitions between distant configurations.
minor comments (2)
  1. [Introduction] Notation for the target Hamiltonian and the auxiliary Ising chain should be introduced with a clear table or diagram distinguishing the two systems.
  2. [Algorithm description] The manuscript would benefit from a short pseudocode block or flowchart summarizing the quantum-classical hybrid loop (Floquet evolution, measurement, acceptance test).

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on our manuscript. We address each major comment below and describe the revisions we will incorporate to strengthen the presentation of the ergodicity and parameter-tuning aspects of the algorithm.

read point-by-point responses

  1. Referee: [Abstract / central construction] Abstract and central construction (method description): The manuscript asserts that MBL properties 'enable one to address the conditions for ergodicity.' However, MBL eigenstates are localized with exponentially decaying correlations and the dynamics are non-ergodic by definition. No explicit argument or derivation is supplied showing that the Floquet evolution in the MBL regime generates a transition kernel that is irreducible over the full configuration space of an arbitrary target Hamiltonian. Without such a demonstration (e.g., a lower bound on the spectral gap or a connectivity argument), the stationary distribution reached by the sampler may be supported only on a restricted subset of states rather than the desired thermal distribution.

    Authors: We appreciate the referee pointing out the need for an explicit demonstration of irreducibility. While MBL dynamics are non-ergodic for the driven chain itself, our QMCMC construction uses the MBL Floquet evolution only as a proposal generator inside a Metropolis-Hastings step whose acceptance probability is chosen to enforce detailed balance with respect to the target Boltzmann distribution. In the revised manuscript we will add a dedicated paragraph in the Methods section that supplies a connectivity argument: the nearest-neighbor Ising Floquet drive, even in the MBL regime, can generate all configurations within a fixed parity sector when the localization length is comparable to system size; combined with the fact that the target Hamiltonian is arbitrary but the proposal kernel has non-zero overlap with every single-spin flip (or small cluster flip), the overall Markov chain is irreducible on the full configuration space. A quantitative lower bound on the spectral gap is left as an open question for future work. revision: yes

  2. Referee: [Method section] Method section (description of the acceptance-rate tuning): The claim that the thermalized-to-localized transition can be exploited to tune the acceptance rate of the Markov chain is load-bearing for practical performance. The manuscript must specify how the localization length or driving parameters are chosen so that the effective proposal distribution remains sufficiently connected while still satisfying detailed balance with respect to the target Boltzmann weight; otherwise the tuning may inadvertently suppress transitions between distant configurations.

    Authors: We agree that concrete guidance on parameter selection is essential for reproducibility and performance. In the revised manuscript we will expand the Methods section to state explicitly that the driving frequency and disorder strength are chosen so that the localization length is of order the system size (i.e., operating near the thermal-to-MBL transition). This choice ensures that the proposal distribution generated by the Floquet evolution permits multi-spin flips with appreciable probability, while the subsequent Metropolis acceptance step is computed exactly from the target Hamiltonian and therefore automatically satisfies detailed balance. We will also include a short numerical illustration showing how the effective acceptance rate varies across the transition for a representative combinatorial optimization instance. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on external MBL phenomenology

full rationale

The paper frames its quantum MCMC construction as utilizing established properties of the many-body localized phase to satisfy ergodicity conditions and tune acceptance rates via the thermalized-to-localized transition. No self-definitional steps, fitted parameters renamed as predictions, or load-bearing self-citations appear in the abstract or described chain. The central claim—that the method enables sampling from thermal distributions of non-native Hamiltonians on programmable simulators of 1D Ising Floquet dynamics—builds on independent MBL literature rather than reducing to its own inputs by construction. This is the expected honest non-finding for a method paper that treats MBL phenomenology as an external input.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the existence and controllable properties of the MBL phase in driven Ising chains and on the ability to map combinatorial problems onto thermal sampling of non-native Hamiltonians.

axioms (2)
  • domain assumption The MBL phase possesses properties that satisfy ergodicity and sampling conditions for quantum Markov chains.
    Invoked in the abstract to justify the algorithm's validity.
  • domain assumption The thermalized-to-localized transition can be used to control Markov chain acceptance rates.
    Central to the tuning mechanism described.

pith-pipeline@v0.9.0 · 5658 in / 1219 out tokens · 45611 ms · 2026-05-19T12:57:39.806532+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

99 extracted references · 99 canonical work pages · 2 internal anchors

  1. [1]

    and high-energy physics [23], but have since paved the way for a plethora of successful Monte Carlo algo- rithms [24–27] that nowadays extend to other fields such as hyperparameter tuning in Bayesian inference [28] and combinatorial optimization through simulated annealing [29, 30]. Given the success and widespread use of Markov chain Monte Carlo methods,...

    work page

  2. [2]

    In a closely related de- velopment, the quantum-enhanced algorithm has been extended to variational Monte Carlo [42]

    has been developed [41]. In a closely related de- velopment, the quantum-enhanced algorithm has been extended to variational Monte Carlo [42]. Another recent work in this hybrid direction [43] proposes a quantum method to improve the classical Langevin and Hamiltonian Monte Carlo. They identified gradient cal- culation as the bottleneck of the algorithm a...

    work page 2000

  3. [3]

    More specifically, letE be the state space of the Markov chain, equipped with a σ-algebraE

    Irreducibility For general state space Markov chains, irreducibility is taken with respect to a measure. More specifically, letE be the state space of the Markov chain, equipped with a σ-algebraE. For anyA∈Eandx∈E, defineP n(x, A) to be the probability that, starting fromx, the Markov chain reaches inniterations a statey∈A. Given aσ- finite measureφsuch t...

    work page

  4. [4]

    the uniformly random measure on the unitary group

    A sufficiently long random chain of MBL unitaries is CUE, i.e. the uniformly random measure on the unitary group

    work page

  5. [5]

    for allx, y∈S 2d−1it exists u∈U(d)such thatu⋅x=y(in our cased=2 N)

    The unitary groupU(d)acts transitively on the hypersphereS 2d−1, i.e. for allx, y∈S 2d−1it exists u∈U(d)such thatu⋅x=y(in our cased=2 N). Therefore, because of (2.) it always exists a unitary that brings fromxtoy, and because of (1.), ifnis sufficiently large, there is a strictly positive probability of that uni- tary being the chain of proposed MBLs. The...

    work page

  6. [6]

    A convenient way of verifying uniform ergodicity is to use the concept of small set

    Uniform ergodicity A Markov chain is uniformly ergodic if there is a positive constantMand 0<r<1 such that: sup x∈E ∣∣Pn(x,⋅)−π∣∣≤M rn (A8) meaning that convergence to the equilibrium distribution happens geometrically fast uniformly inx. A convenient way of verifying uniform ergodicity is to use the concept of small set. A subsetC⊂Eis(n 0, ϵ, ν)-small if...

    work page

  7. [7]

    Bounds on mixing time Bounds on convergence can be formulated in terms of the acceptance probabilityT(x): T(x)∶= ∫E q(x, y)α(x, y)µ L(dy).(A10) For uniformly ergodic chains, Eq. (A8), one has ([70] Lemma 3): r≥1−T(x).(A11) Moreover, defining the mixing timeτ ϵ as the number of iterations required to reach a certain error thresholdϵ: τϵ ∶=inf n∈Z+ [sup x∈E...

    work page

  8. [8]

    Therefore, all other pa- rameters being fixed,Wshould be chosen so that the acceptance rate of the Markov chain is the desired one

    Disorder strengthW We showed in the main text howWis essentially the one parameter controlling the acceptance rate of the Markov chain, as it is inherently related to the transition between thermalization and localization. Therefore, all other pa- rameters being fixed,Wshould be chosen so that the acceptance rate of the Markov chain is the desired one. In...

    work page

  9. [9]

    However, the choice ofωand Jholds an important practical value

    Drive frequencyωand nearest-neighbour interactionJ On paper, the one important parameter of Metropolis is the acceptance rate, and therefore one could argue thatω andJcan be set almost arbitrarily, as long as a suitable value forWcan then be found to bring the acceptance rate to the desired value. However, the choice ofωand Jholds an important practical v...

    work page

  10. [10]

    2 of [53], in order to make sure that the parameters result in an MBL unitary

    Drive amplitudeδBandB 0 When tuning parameters likeJandωfor experimen- tal purposes, one should always keep in mind the thermalization-localization phase diagram, as shown in Fig. 2 of [53], in order to make sure that the parameters result in an MBL unitary. The diagram depends only on the dimensionless quantitiesω/JandW/J, which leaves the choice ofJfree...

    work page 2000

  11. [11]

    Laser intensity, frequency, and phase fluctuations affecting Ω(t)andδ i

    work page

  12. [12]

    Atomic position errors affectinga

    work page

  13. [13]

    Coupling with the environment

    work page

  14. [14]

    SPAM errors 14 Laser fluctuations are in general of two types: high fre- quency (affecting the intensity and detuning within a sin- gle shot) and low frequency (causing shot-to-shot varia- tions in intensity and detuning). High-precision numeri- cal simulations [82] show how the decoherence induced by such noise sources is somewhat suppressed by operating...

    work page

  15. [15]

    , N}and edge setE⊆V×V

    MIS LetG=(V, E)be a graph of sizeNdefined by its vertex setV={1, . . . , N}and edge setE⊆V×V. Let ⃗x= (x1, . . . , xN)∈{0,1}N be a binary vector of lengthN. A solution to the Maximum Independent Set problem is given by any vector ⃗x∗that minimizes the following quadratic cost function: ⃗x∗=argmin ⃗x∈{0,1}N ⎛ ⎝− ∑ i∈V xi +2 ∑ (i,j)∈E xixj ⎞ ⎠.(F1) Given a ...

    work page

  16. [16]

    MAX-CUT LetG=(V, E)be a graph as above. A solution to the Maximum Cut problem is given by any vector ⃗x∗that minimizes: ⃗x∗=argmin ⃗x∈{0,1}N ⎛ ⎝− ∑ i∈V Nixi +2 ∑ (i,j)∈E xixj ⎞ ⎠.(F2) WhereN i =∣{j∈V∶(i, j)∈E}∣is the degree (number of neighbours) of vertexi

    work page

  17. [17]

    LetNbe an integer that we wish to factorize into two primespandq

    Prime Factorization This derivation follows [60]. LetNbe an integer that we wish to factorize into two primespandq. The prime factorization problem can be formulated as a least-squares problem by minimizing the difference between the productpqandN: argmin p,q∈N (pq−N) 2.(F3) The minimum of this expression is 0, and it is achieved whenpq=N. Expanding the s...

    work page

  18. [18]

    Mostame, J

    S. Mostame, J. Huh, C. Kreisbeck, A. J. Kerman, T. Fu- jita, A. Eisfeld, and A. Aspuru-Guzik, Quantum Inf. Process.16, 44 (2016)

    work page 2016

  19. [19]

    M. W. Johnson, M. H. S. Amin, S. Gildert, T. Lanting, F. Hamze, N. Dickson, R. Harris, A. J. Berkley, J. Jo- hansson, P. Bunyk, E. M. Chapple, C. Enderud, J. P. Hilton, K. Karimi, E. Ladizinsky, N. Ladizinsky, T. Oh, I. Perminov, C. Rich, M. C. Thom, E. Tolkacheva, C. J. S. 15 Truncik, S. Uchaikin, J. Wang, B. Wilson, and G. Rose, Nature473, 194 (2011)

    work page 2011

  20. [20]

    Computational supremacy in quantum simulation,

    A. D. King, A. Nocera, M. M. Rams, J. Dziarmaga, R. Wiersema, W. Bernoudy, J. Raymond, N. Kaushal, N. Heinsdorf, R. Harris, K. Boothby, F. Altomare, A. J. Berkley, M. Boschnak, K. Chern, H. Chris- tiani, S. Cibere, J. Connor, M. H. Dehn, R. Desh- pande, S. Ejtemaee, P. Farr´ e, K. Hamer, E. Hoskin- son, S. Huang, M. W. Johnson, S. Kortas, E. Ladizin- sky,...

    work page arXiv 2024

  21. [21]

    Semeghini, H

    G. Semeghini, H. Levine, A. Keesling, S. Ebadi, T. T. Wang, D. Bluvstein, R. Verresen, H. Pich- ler, M. Kalinowski, R. Samajdar, A. Omran, S. Sachdev, A. Vishwanath, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Science374, 1242 (2021), https://www.science.org/doi/pdf/10.1126/science.abi8794

    work page doi:10.1126/science.abi8794 2021

  22. [22]

    Aaronson, Theory of Computing Systems55, 281 (2014)

    S. Aaronson, Theory of Computing Systems55, 281 (2014)

    work page 2014

  23. [23]

    Tangpanitanon, S

    J. Tangpanitanon, S. Thanasilp, N. Dangniam, M.-A. Lemonde, and D. G. Angelakis, Phys. Rev. Res.2, 043364 (2020)

    work page 2020

  24. [24]

    V. M. Bastidas, T. Haug, C. Gravel, L.-C. Kwek, W. J. Munro, and K. Nemoto, Phys. Rev. B105, 075140 (2022)

    work page 2022

  25. [25]

    Rigol, V

    M. Rigol, V. Dunjko, and M. Olshanii, Nature452, 854 (2008)

    work page 2008

  26. [26]

    P. W. Anderson, Phys. Rev.109, 1492 (1958)

    work page 1958

  27. [27]

    D. M. Basko, I. L. Aleiner, and B. L. Altshuler, Ann. Phys.321, 1126 (2006)

    work page 2006

  28. [28]

    Oganesyan and D

    V. Oganesyan and D. A. Huse, Phys. Rev. B75, 155111 (2007)

    work page 2007

  29. [29]

    Nandkishore and D

    R. Nandkishore and D. A. Huse, Annu. Rev. Condens. Matter Phys.6, 15 (2015)

    work page 2015

  30. [30]

    D. A. Abanin, W. De Roeck, and F. Huveneers, Annals of Physics372, 1 (2016)

    work page 2016

  31. [31]

    Serbyn, Z

    M. Serbyn, Z. Papic, and D. A. Abanin, Phys. Rev. Lett. 111, 127201 (2013)

    work page 2013

  32. [32]

    D. A. Huse, R. Nandkishore, and V. Oganesyan, Phys. Rev. B90, 174202 (2014)

    work page 2014

  33. [33]

    J. Z. Imbrie, V. Ros, and A. Scardicchio, Ann. Phys. (Berlin)529, 1600278 (2017)

    work page 2017

  34. [34]

    Metropolis, A

    N. Metropolis, A. W. Rosenbluth, M. N. Rosenbluth, A. H. Teller, and E. Teller, The Journal of Chemical Physics21, 1087 (1953), https://pubs.aip.org/aip/jcp/article- pdf/21/6/1087/18802390/1087 1 online.pdf

    work page 1953

  35. [35]

    W. K. Hastings, Biometrika57, 97 (1970)

    work page 1970

  36. [36]

    Optimal conversion from classical to quantum randomness via quantum chaos,

    W.-K. Mok, T. Haug, A. L. Shaw, M. Endres, and J. Preskill, “Optimal conversion from classical to quantum randomness via quantum chaos,” (2025), arXiv:2410.05181 [quant-ph]

    work page arXiv 2025

  37. [37]

    J. R. Norris,Markov Chains, Cambridge Series in Statis- tical and Probabilistic Mathematics (Cambridge Univer- sity Press, Cambridge, 1997)

    work page 1997

  38. [38]

    Krauth,Statistical Mechanics: Algorithms and Com- putations(Oxford University Press, Oxford, 2006)

    W. Krauth,Statistical Mechanics: Algorithms and Com- putations(Oxford University Press, Oxford, 2006)

    work page 2006

  39. [39]

    M. E. J. Newman and G. T. Barkema,Monte Carlo Methods in Statistical Physics(Oxford University Press, U.S.A., 1999) p. 496, illustrated edition, published April 15

    work page 1999

  40. [40]

    Montvay and G

    I. Montvay and G. M¨ unster,Quantum Fields on a Lat- tice, Cambridge Monographs on Mathematical Physics (Cambridge University Press, 1994)

    work page 1994

  41. [41]

    G. O. Roberts and R. L. Tweedie, Bernoulli2, 341 (1996)

    work page 1996

  42. [42]

    Duane, A

    S. Duane, A. Kennedy, B. J. Pendleton, and D. Roweth, Physics Letters B195, 216 (1987)

    work page 1987

  43. [43]

    R. H. Swendsen and J.-S. Wang, Phys. Rev. Lett.57, 2607 (1986)

    work page 1986

  44. [44]

    Marinari and G

    E. Marinari and G. Parisi, Europhysics Letters19, 451 (1992)

    work page 1992

  45. [45]

    M. D. Homan and A. Gelman, J. Mach. Learn. Res.15, 1593–1623 (2014)

    work page 2014

  46. [46]

    Kirkpatrick, C

    S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, Science220, 671 (1983), https://www.science.org/doi/pdf/10.1126/science.220.4598.671

    work page doi:10.1126/science.220.4598.671 1983

  47. [47]

    Martoˇ n´ ak, G

    R. Martoˇ n´ ak, G. E. Santoro, and E. Tosatti, Phys. Rev. B66, 094203 (2002)

    work page 2002

  48. [48]

    Szegedy, in45th Annual IEEE Symposium on Foun- dations of Computer Science(2004) pp

    M. Szegedy, in45th Annual IEEE Symposium on Foun- dations of Computer Science(2004) pp. 32–41

    work page 2004

  49. [49]

    Walking through hilbert space with quantum computers,

    T. Jiang, J. Zhang, M. K. A. Baumgarten, M.-F. Chen, H. Q. Dinh, A. Ganeshram, N. Maskara, A. Ni, and J. Lee, “Walking through hilbert space with quantum computers,” (2024), arXiv:2407.11672 [quant-ph]

    work page arXiv 2024

  50. [50]

    Temme, T

    K. Temme, T. J. Osborne, K. G. Vollbrecht, D. Poulin, and F. Verstraete, Nature471, 87 (2011)

    work page 2011

  51. [51]

    Quantum Thermal State Preparation

    C.-F. Chen, M. J. Kastoryano, F. G. S. L. Brand˜ ao, and A. Gily´ en, “Quantum thermal state preparation,” (2023), arXiv:2303.18224 [quant-ph]

    work page internal anchor Pith review arXiv 2023

  52. [52]

    Marriott and J

    C. Marriott and J. Watrous, computational complexity 14, 122 (2005)

    work page 2005

  53. [53]

    Quantum metropolis sampling via weak measurement,

    J. Jiang and S. Irani, “Quantum metropolis sampling via weak measurement,” (2024), arXiv:2406.16023 [quant- ph]

    work page arXiv 2024

  54. [54]

    An efficient and exact noncommutative quantum Gibbs sampler.arXiv:2311.09207, 2023

    C.-F. Chen, M. J. Kastoryano, and A. Gily´ en, “An ef- ficient and exact noncommutative quantum gibbs sam- pler,” (2023), arXiv:2311.09207 [quant-ph]

    work page arXiv 2023

  55. [55]

    Quantum generalizations of glauber and metropolis dynamics,

    A. Gily´ en, C.-F. Chen, J. F. Doriguello, and M. J. Kastoryano, “Quantum generalizations of glauber and metropolis dynamics,” (2024), arXiv:2405.20322 [quant- ph]

    work page arXiv 2024

  56. [56]

    Preparing thermal states on noiseless and noisy programmable quantum processors

    O. Shtanko and R. Movassagh, “Preparing thermal states on noiseless and noisy programmable quantum proces- sors,” (2023), arXiv:2112.14688 [quant-ph]

    work page arXiv 2023

  57. [57]

    Layden, G

    D. Layden, G. Mazzola, R. V. Mishmash, R. T. Babbush, S. Bravyi, A. W. Harrow, J. A. Gross, R. Sweke, J. I. Cirac, J. M. Gambetta, and J. M. Epstein, Nature619, 282 (2023)

    work page 2023

  58. [58]

    From quantum enhanced to quantum inspired monte carlo,

    J. Christmann, P. Ivashkov, M. Chiurco, and G. Maz- zola, “From quantum enhanced to quantum inspired monte carlo,” (2024), arXiv:2411.17821 [quant-ph]

    work page arXiv 2024

  59. [59]

    Quantum-assisted variational monte carlo,

    L. Chang, Z. Li, and W.-H. Fang, “Quantum-assisted variational monte carlo,” (2025), arXiv:2502.20799 [quant-ph]

    work page arXiv 2025

  60. [60]

    Quantum speedups for markov chain monte carlo methods with application to optimization,

    G. Ozgul, X. Li, M. Mahdavi, and C. Wang, “Quantum speedups for markov chain monte carlo methods with application to optimization,” (2025), arXiv:2504.03626 [quant-ph]

    work page arXiv 2025

  61. [61]

    G. O. Roberts, A. Gelman, and W. R. Gilks, The Annals of Applied Probability7, 110 (1997)

    work page 1997

  62. [62]

    Breyer and G

    L. Breyer and G. Roberts, Stochastic Processes and their Applications90, 181 (2000). 16

    work page 2000

  63. [63]

    G. O. Roberts and J. S. Rosenthal, Statistical Science 16, 351 (2001)

    work page 2001

  64. [64]

    B´ edard, Stochastic Processes and their Applications 118, 2198 (2008)

    M. B´ edard, Stochastic Processes and their Applications 118, 2198 (2008)

    work page 2008

  65. [65]

    Rao, The European Physical Journal Plus137, 398 (2022)

    W.-J. Rao, The European Physical Journal Plus137, 398 (2022)

    work page 2022

  66. [66]

    Haake, S

    F. Haake, S. Gnutzmann, and M. Ku´ s,Quantum Signa- tures of Chaos(Springer Cham, 2018)

    work page 2018

  67. [67]

    Ponte, Z

    P. Ponte, Z. Papic, F. Huveneers, and D. A. Abanin, Phys. Rev. Lett.114, 140401 (2015)

    work page 2015

  68. [68]

    Lazarides, A

    A. Lazarides, A. Das, and R. Moessner, Phys. Rev. Lett. 115, 030402 (2015)

    work page 2015

  69. [69]

    D. A. Abanin, W. De Roeck, and F. Huveneers, Ann. Phys. (Berlin)529, 1600182 (2017)

    work page 2017

  70. [70]

    Thanasilp, J

    S. Thanasilp, J. Tangpanitanon, M.-A. Lemonde, N. Dangniam, and D. G. Angelakis, Phys. Rev. B103, 165132 (2021)

    work page 2021

  71. [71]

    Propaga- tion of errors and quantitative quantum simulation with quantum advantage,

    S. Flannigan, N. Pearson, G. H. Low, A. Buyskikh, I. Bloch, P. Zoller, M. Troyer, and A. J. Daley, “Propaga- tion of errors and quantitative quantum simulation with quantum advantage,” (2022), arXiv:2204.13644 [quant- ph]

    work page arXiv 2022

  72. [72]

    Gonzalez-Conde, Z

    J. Gonzalez-Conde, Z. Morrell, M. Vuffray, T. Albash, and C. Coffrin, Phys. Rev. A111, 062606 (2025)

    work page 2025

  73. [73]

    The quantum adiabatic algorithm suppresses the prolif- eration of errors,

    B. F. Schiffer, A. F. Rubio, R. Trivedi, and J. I. Cirac, “The quantum adiabatic algorithm suppresses the prolif- eration of errors,” (2024), arXiv:2404.15397 [quant-ph]

    work page arXiv 2024

  74. [74]

    Jun and H

    K. Jun and H. Lee, Results in Control and Optimization 11, 100222 (2023)

    work page 2023

  75. [75]

    Boros and P

    E. Boros and P. L. Hammer, Discrete Applied Mathe- matics123, 155 (2002)

    work page 2002

  76. [76]

    Computers, complex- ity, and intractability from the parametric point of view,

    R. G. Downey and M. R. Fellows, “Computers, complex- ity, and intractability from the parametric point of view,” inParameterized Complexity(Springer New York, New York, NY, 1999) pp. 1–20

    work page 1999

  77. [77]

    Jun and H

    K. Jun and H. Lee, Scientific Reports13, 10080 (2023)

    work page 2023

  78. [78]

    Mandal, A

    A. Mandal, A. Roy, S. Upadhyay, and H. Ushijima- Mwesigwa, in2020 Data Compression Conference (DCC) (2020) pp. 383–383

    work page 2020

  79. [79]

    K. Mato, R. Mengoni, D. Ottaviani, and G. Palermo, Quantum Science and Technology7, 035020 (2022)

    work page 2022

  80. [80]

    Pelofske, G

    E. Pelofske, G. Hahn, D. O’Malley, H. N. Djidjev, and B. S. Alexandrov, Scientific Reports12, 8539 (2022)

    work page 2022

Showing first 80 references.