- Physical Review Journals
All JournalsPhysics Magazine
Access Provided by Xinjiang University

Physical Review A

Export citation

Export citation

Choose format for download:

Download Citation
  • Access by Xinjiang University

Comparing quantum to simulated reverse annealing in mean-field models

Christopher L. Baldwin*

  • *Contact author: baldw292@msu.edu
PDF

Phys. Rev. A 114, 022422 – Published 12 August, 2026

DOI: https://doi.org/10.1103/ddv6-66hd

Abstract

Adiabatic reverse annealing (ARA) has been proposed as an improvement to conventional quantum annealing for solving optimization problems, in which one takes advantage of an initial guess at the solution to suppress problematic phase transitions. Here we interpret the performance of ARA through its effects on the free-energy landscape and use the intuition gained to introduce a classical analog to ARA termed “simulated reverse annealing” (SRA). The close similarity between the two means that they should be compared side by side in applications since SRA could potentially give comparable results. As a solvable example, we analyze how both protocols behave in a family of infinite-range (nondisordered) p-spin models. Through both the thermodynamic phase diagrams and explicit dynamical behavior, we establish that the quantum algorithm has no advantage over its classical counterpart (in the range of models considered): SRA succeeds at avoiding discontinuous transitions not only in every case where ARA does, but even in a narrow window of parameters where ARA fails to do so.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (64)

  1. T. Albash and D. A. Lidar, Adiabatic quantum computation, Rev. Mod. Phys. 90, 015002 (2018).
  2. P. Hauke, H. G. Katzgraber, W. Lechner, H. Nishimori, and W. D. Oliver, Perspectives of quantum annealing: Methods and implementations, Rep. Prog. Phys. 83, 054401 (2020).
  3. A. Rajak, S. Suzuki, A. Dutta, and B. K. Chakrabarti, Quantum annealing: An overview, Phil. Trans. R. Soc. A 381, 20210417 (2023).
  4. A. Messiah, Quantum Mechanics (North-Holland, Amsterdam, 1962), Vol. II.
  5. S. Jansen, M.-B. Ruskai, and R. Seiler, Bounds for the adiabatic approximation with applications to quantum computation, J. Math. Phys. 48, 102111 (2007).
  6. M. E. J. Newman and G. T. Barkema, Monte Carlo Methods in Statistical Physics (Clarendon Press, Oxford, UK, 1999).
  7. T. Jörg, F. Krzakala, J. Kurchan, and A. C. Maggs, Simple glass models and their quantum annealing, Phys. Rev. Lett. 101, 147204 (2008).
  8. B. Altshuler, H. Krovi, and J. Roland, Anderson localization makes adiabatic quantum optimization fail, Proc. Natl. Acad. Sci. USA 107, 12446 (2010).
  9. T. Jörg, F. Krzakala, G. Semerjian, and F. Zamponi, First-order transitions and the performance of quantum algorithms in random optimization problems, Phys. Rev. Lett. 104, 207206 (2010).
  10. A. P. Young, S. Knysh, and V. N. Smelyanskiy, First-order phase transition in the quantum adiabatic algorithm, Phys. Rev. Lett. 104, 020502 (2010).
  11. I. Hen and A. P. Young, Exponential complexity of the quantum adiabatic algorithm for certain satisfiability problems, Phys. Rev. E 84, 061152 (2011).
  12. E. Farhi, D. Gosset, I. Hen, A. W. Sandvik, P. Shor, A. P. Young, and F. Zamponi, Performance of the quantum adiabatic algorithm on random instances of two optimization problems on regular hypergraphs, Phys. Rev. A 86, 052334 (2012).
  13. V. Bapst, L. Foini, F. Krzakala, G. Semerjian, and F. Zamponi, The quantum adiabatic algorithm applied to random optimization problems: The quantum spin glass perspective, Phys. Rep. 523, 127 (2013).
  14. S. Knysh, Zero-temperature quantum annealing bottlenecks in the spin-glass phase, Nat. Commun. 7, 12370 (2016).
  15. C. L. Baldwin and C. R. Laumann, Quantum algorithm for energy matching in hard optimization problems, Phys. Rev. B 97, 224201 (2018).
  16. Y. Seki and H. Nishimori, Quantum annealing with antiferromagnetic fluctuations, Phys. Rev. E 85, 051112 (2012).
  17. L. Hormozi, E. W. Brown, G. Carleo, and M. Troyer, Nonstoquastic Hamiltonians and quantum annealing of an Ising spin glass, Phys. Rev. B 95, 184416 (2017).
  18. Y. Susa, Y. Yamashiro, M. Yamamoto, and H. Nishimori, Exponential speedup of quantum annealing by inhomogeneous driving of the transverse field, J. Phys. Soc. Jpn. 87, 023002 (2018).
  19. Y. Susa, Y. Yamashiro, M. Yamamoto, I. Hen, D. A. Lidar, and H. Nishimori, Quantum annealing of the p-spin model under inhomogeneous transverse field driving, Phys. Rev. A 98, 042326 (2018).
  20. T. Albash, Role of nonstoquastic catalysts in quantum adiabatic optimization, Phys. Rev. A 99, 042334 (2019).
  21. J. I. Adame and P. L. McMahon, Inhomogeneous driving in quantum annealers can result in orders-of-magnitude improvements in performance, Quantum Sci. Technol. 5, 035011 (2020).
  22. Z. Tang and E. Kapit, Unconventional quantum annealing methods for difficult trial problems, Phys. Rev. A 103, 032612 (2021).
  23. T. Imoto, Y. Seki, Y. Matsuzaki, and S. Kawabata, Quantum annealing with twisted fields, New J. Phys. 24, 113009 (2022).
  24. Y. Kimura and H. Nishimori, Convergence condition of simulated quantum annealing with a non-stoquastic catalyst, J. Phys. A: Math. Theor. 56, 165304 (2023).
  25. M. Dadgar and C. L. Baldwin, Anomalously slow dynamics of inhomogeneous quantum annealing, Phys. Rev. A 111, 062601 (2025).
  26. A. Perdomo-Ortiz, S. E. Venegas-Andraca, and A. Aspuru-Guzik, A study of heuristic guesses for adiabatic quantum computation, Quantum Inf. Proc. 10, 33 (2011).
  27. N. Chancellor, Modernizing quantum annealing using local searches, New J. Phys. 19, 023024 (2017).
  28. J. Marshall, D. Venturelli, I. Hen, and E. G. Rieffel, Power of pausing: Advancing understanding of thermalization in experimental quantum annealers, Phys. Rev. Appl. 11, 044083 (2019).
  29. G. Passarelli, K.-W. Yip, D. A. Lidar, H. Nishimori, and P. Lucignano, Reverse quantum annealing of the p-spin model with relaxation, Phys. Rev. A 101, 022331 (2020).
  30. L. Rocutto, C. Destri, and E. Prati, Quantum semantic learning by reverse annealing of an adiabatic quantum computer, Adv. Quantum Technol. 4, 2000133 (2021).
  31. Y. Bando, K.-W. Yip, H. Chen, D. A. Lidar, and H. Nishimori, Breakdown of the weak-coupling limit in quantum annealing, Phys. Rev. Appl. 17, 054033 (2022).
  32. V. Mehta, H. De Raedt, K. Michielsen, and F. Jin, Unraveling reverse annealing: A study of D-Wave quantum annealers, Phys. Rev. A 112, 012414 (2025).
  33. M. Ohkuwa, H. Nishimori, and D. A. Lidar, Reverse annealing for the fully connected p-spin model, Phys. Rev. A 98, 022314 (2018).
  34. Y. Yamashiro, M. Ohkuwa, H. Nishimori, and D. A. Lidar, Dynamics of reverse annealing for the fully connected p-spin model, Phys. Rev. A 100, 052321 (2019).
  35. S. Arai, M. Ohzeki, and K. Tanaka, Mean field analysis of reverse annealing for code-division multiple-access multiuser detection, Phys. Rev. Res. 3, 033006 (2021).
  36. S. Arai, Mean-field analysis of Sourlas codes with adiabatic reverse annealing, in Sublinear Computation Paradigm, edited by N. Katoh, Y. Higashikawa, H. Ito, A. Nagao, T. Shibuya, A. Sljoka, K. Tanaka, and Y. Uno (Springer, New York, 2022), Chap. 13, pp. 319.
  37. G. Passarelli and P. Lucignano, Counterdiabatic reverse annealing, Phys. Rev. A 107, 022607 (2023).
  38. H. Zhang, K. Boothby, and A. Kamenev, Cyclic quantum annealing: Searching for deep low-energy states in 5000-qubit spin glass, Sci. Rep. 14, 30784 (2024).
  39. M. W. Johnson, M. H. S. Amin, S. Gildert, T. Lanting, F. Hamze, N. Dickson, R. Harris, A. J. Berkley, J. Johansson, 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, et al., Quantum annealing with manufactured spins, Nature (London) 473, 194 (2011).
  40. A. D. King, S. Suzuki, J. Raymond, A. Zucca, T. Lanting, F. Altomare, A. J. Berkley, S. Ejtemaee, E. Hoskinson, S. Huang, E. Ladizinsky, A. J. R. MacDonald, G. Marsden, T. Oh, G. Poulin-Lamarre, M. Reis, C. Rich, Y. Sato, J. D. Whittaker, J. Yao, et al., Coherent quantum annealing in a programmable 2,000 qubit Ising chain, Nat. Phys. 18, 1324 (2022).
  41. A. D. King, J. Raymond, T. Lanting, R. Harris, A. Zucca, F. Altomare, A. J. Berkley, K. Boothby, S. Ejtemaee, C. Enderud, E. Hoskinson, S. Huang, E. Ladizinsky, A. J. R. MacDonald, G. Marsden, R. Molavi, T. Oh, G. Poulin-Lamarre, M. Reis, C. Rich, et al., Quantum critical dynamics in a 5,000-qubit programmable spin glass, Nature (London) 617, 61 (2023).
  42. D. J. Amit, H. Gutfreund, and H. Sompolinsky, Spin-glass models of neural networks, Phys. Rev. A 32, 1007 (1985).
  43. H. Nishimori and Y. Nonomura, Quantum effects in neural networks, J. Phys. Soc. Jpn. 65, 3780 (1996).
  44. S. W. Shin, G. Smith, J. A. Smolin, and U. Vazirani, How “quantum” is the D-Wave machine? arXiv:1401.7087.
  45. T. Albash and J. Marshall, Comparing relaxation mechanisms in quantum and classical transverse-field annealing, Phys. Rev. Appl. 15, 014029 (2021).
  46. D. Subires, F. J. Gómez-Ruiz, A. Ruiz-García, D. Alonso, and A. del Campo, Benchmarking quantum annealing dynamics: The spin-vector Langevin model, Phys. Rev. Res. 4, 023104 (2022).
  47. G. Passarelli, K.-W. Yip, D. A. Lidar, and P. Lucignano, Standard quantum annealing outperforms adiabatic reverse annealing with decoherence, Phys. Rev. A 105, 032431 (2022).
  48. T. Jörg, F. Krzakala, J. Kurchan, A. C. Maggs, and J. Pujos, Energy gaps in quantum first-order mean-field-like transitions: The problems that quantum annealing cannot solve, Europhys. Lett. 89, 40004 (2010).
  49. V. Bapst and G. Semerjian, On quantum mean-field models and their quantum annealing, J. Stat. Mech.: Theory Expt. (2012) P06007.
  50. The static ansatz is equivalently the assumption that imaginary-time-translation symmetry is not spontaneously broken, which is a very reasonable assumption to make.
  51. One can see this by following the same derivation of the path integral, but for the thermal expectation value of ∑jδaj,1σ̂jz/N (similarly for ∑jδaj,−1σ̂jz/N) rather than the partition function. The result is the path integral of muexp(−NβΦ)/Z, which according to the saddle-point approximation is simply mu evaluated at the saddle point [the factor of exp(−NβΦ) cancels with Z in the denominator].
  52. There are subtleties in defining the free-energy landscape. For one thing, the fields (hu,hd) should not be treated as additional variables in a four-dimensional landscape, but rather as functions of (mu,md) determined by setting ∂Φ/∂hu and ∂Φ/∂hd to zero. Furthermore, it is important to take the correct derivatives—even though setting ∂Φ/∂mu=0 immediately gives hu as a function of (mu,md), this is the wrong function to use. Further discussion can be found in the Appendix of Ref. [64].
  53. For this reason, we refer to Φ(mu,md) as the free-energy landscape even at T=0—although there are no thermal fluctuations, the transverse field has a very similar effect. For our purposes, it is more convenient to think of the energy landscape as that of H0 alone (for which the local minima are exactly at the all-up and marked states), and consider any fluctuations as giving rise to a free-energy landscape.
  54. We determine the locations of the discontinuous phase transitions simply by where (mu,md) changes by more than 0.05 (in two-dimensional Euclidean distance) when increasing s from one pixel to the next (using a resolution in s of Δs=0.002). This is admittedly crude, but note that the dynamical calculations do not require any knowledge of the phase boundaries—we include them in the insets merely to help visualize the protocol paths.
  55. This statement is more subtle than it may appear: even though both local minima are stable on either side of a discontinuous phase transition, there can be separate “spinodal” lines at which the secondary minimum becomes unstable. If the spinodal is crossed, there is no longer a (free-) energy barrier and the system may be able to reach the desired ground state, especially via SRA since it directly moves downhill in free energy. The figure uses parameter values for which there is no such spinodal.
  56. A. Le and C. L. Baldwin, Adiabatic reverse annealing is robust to low-temperature decoherence, arXiv:2511.16735.
  57. L. F. Cugliandolo and G. Lozano, Real-time nonequilibrium dynamics of quantum glassy systems, Phys. Rev. B 59, 915 (1999).
  58. T. Castellani and A. Cavagna, Spin-glass theory for pedestrians, J. Stat. Mech. (2005) P05012.
  59. M. Mezard and A. Montanari, Information, Physics, and Computation (Oxford University Press, Oxford, 2009).
  60. S. Franz and G. Parisi, Recipes for metastable states in spin glasses, J. Phys. I France 5, 1401 (1995).
  61. H. F. Trotter, On the product of semi-groups of operators, Proc. Am. Math. Soc. 10, 545 (1959).
  62. M. Suzuki, Relationship between d-dimensional quantal spin systems and (d+1)-dimensional Ising systems: Equivalence, critical exponents and systematic approximants of the partition function and spin correlations, Prog. Theor. Phys. 56, 1454 (1976).
  63. S. Sachdev, Quantum Phase Transitions (Cambridge University Press, Cambridge, UK, 2011).
  64. C. L. Baldwin and B. Swingle, Revisiting the replica trick: Competition between spin glass and conventional order, J. Stat. Phys. 190, 125 (2023).

Outline

Information

Phys. Rev. A 114, 022422– Published 12 August, 2026
  • Received 14 January 2026
  • Accepted 17 July 2026
Article part of CHORUS
Crossmark - Check for updates button

©2026 American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation