- Physical Review Journals
All JournalsPhysics Magazine

Physical Review Letters

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access

Physics of Local Optimization in Complex Disordered Systems

Mutian Shen1,2, Gerardo Ortiz3,4, Zhiqiao Dong5, Martin Weigel6,7, and Zohar Nussinov1,6,8,9,*

  • *Contact author: zohar@wustl.edu
PDF

Phys. Rev. Lett. 137, 167401 – Published 14 October, 2026

DOI: https://doi.org/10.1103/kcbw-6zkv

Abstract

Limited resources motivate decomposing large-scale problems into smaller, “local” subsystems and stitching together the so-found solutions. We explore the physics underlying this approach and discuss the concept of “local hardness,” i.e., the complexity of predicting local properties of the solution from local information, for the ground-state problem of both P- and NP-hard spin glasses and related frustrated spin systems. Depending on the model considered, we observe varying scaling behaviors in how errors associated with local predictions decay as a function of the size of the solved subsystem. These errors are intimately connected to global critical threshold instabilities, characterized by gapless, avalanchelike excitations that follow scale-invariant size distributions. Away from criticality, local solvers quickly achieve high accuracy, aligning closely with the results of the computationally much more expensive global minimization. We leverage these findings to introduce a heuristic contraction-based algorithm for globally studying spin-glass ground states. The local solvers further display sharp imprints of the phase transition from the spin-glass to the ferromagnetic phase as the distribution of spin-glass couplings is shifted, as well as characteristic differences for the infinite-range model, implying the existence of specific classes of local hardness. Our findings shed light on how nature may operate solely through local actions at her disposal.

View figure in article

Physics Subject Headings (PhySH)

Article Text

Supplemental Material

References (72)

  1. A. Krizhevsky, I. Sutskever, and G. E. Hinton, Imagenet classification with deep convolutional neural networks, in Advances in Neural Information Processing Systems (NeurIPS), edited by F. Pereira, C. J. Burges, L. Bottou, and K. Weinberger (Curran Associates, New York, 2012).
  2. J. Jumper et al., Highly accurate protein structure prediction with AlphaFold, Nature (London) 596, 583 (2021).
  3. W. Kohn, Density functional and density matrix method scaling linearly with the number of atoms, Phys. Rev. Lett. 76, 3168 (1996).
  4. D. Ruelle, Cluster property of the correlation functions of classical gases, Rev. Mod. Phys. 36, 580 (1964).
  5. E. H. Lieb and D. W. Robinson, The finite group velocity of quantum spin systems, Commun. Math. Phys. 28, 251 (1972).
  6. B. Nachtergaele and R. Sims, Lieb-Robinson bounds and the exponential clustering theorem, Commun. Path. Phys. 265, 119 (2006).
  7. M. Hastings and T. Koma, Spectral gap and exponential decay of correlations, Commun. Math. Phys. 265, 781 (2006).
  8. B. A. Cipra, The Ising model is NP-complete, SIAM News 33, 6 (2000).
  9. F. Barahona, On the computational complexity of Ising spin glass models, J. Phys. A 15, 3241 (1982).
  10. M. Mezard, G. Parisi, and M. A. Virasoro, Spin Glass Theory and Beyond (World Scientific, Singapore, 1987).
  11. D. L. Stein and C. M. Newman, Spin Glasses and Complexity (Princeton University Press, Princeton, New Jersey, 2013).
  12. F. Menczer, S. Fortunato, and C. A. Davis, A First Course in Network Science (Cambridge University Press, Cambridge, England, 2020).
  13. M. Newman, Networks: An Introduction (Oxford University Press, Oxford, 2018), 2nd ed.
  14. J. D. Bryngelson and P. G. Wolynes, Spin glasses and the statistical mechanics of protein folding, Proc. Natl. Acad. Sci. U.S.A. 84, 7425 (1987).
  15. Rugged Free Energy Landscapes—Common Computational Approaches to Spin Glasses, Structural Glasses and Biological Macromolecules, Lect. Notes Phys. edited by W. Janke (Springer, Berlin, 2007), Vol. 736.
  16. J. Lamarcq, J. P. Bouchaud, O. C. Martin, and M. Mézard, Non-compact local excitations in spin-glasses, Europhys. Lett. 58, 321 (2002).
  17. F. Krzakala and J.-P. Bouchaud, Disorder chaos in spin glasses, Europhys. Lett. 72, 472 (2005).
  18. M. Shen, G. Ortiz, Y.-Y. Liu, M. Weigel, and Z. Nussinov, Universal fragility of spin-glass ground states under single bond changes, Phys. Rev. Lett. 132, 247101 (2024).
  19. C. H. Papadimitriou, Computational Complexity (Addison-Wesley, Reading, MA, 1994).
  20. L. Bieche, J. P. Uhry, R. Maynard, and R. Rammal, On the ground-states of the frustration model of a spin-glass by a matching method of graph-theory, J. Phys. A 13, 2553 (1980).
  21. S. F. Edwards and P. W. Anderson, Theory of spin glasses, J. Phys. F 5, 965 (1975).
  22. D. Sherrington and S. Kirkpatrick, Solvable model of a spin-glass, Phys. Rev. Lett. 35, 1792 (1975).
  23. B. Sun, B. Leonard, P. Ronhovde, and Z. Nussinov, An interacting replica approach applied to the traveling salesman problem, 2016 SAI Computing Conference (IEEE, New York, 2016), p. 319.
  24. W. Wang, S. Mandrà, and H. G. Katzgraber, Patch-planting spin-glass solution for benchmarking, Phys. Rev. E 96, 023312 (2017).
  25. C. Fan, M. Shen, Z. Nussinov, Z. Liu, Y. Sun, and Y.-Y. Liu, Finding spin glass ground states through deep reinforcement learning, Nat. Commun. 14, 725 (2023).
  26. C. Fan, M. Shen, Z. Nussinov, Z. Liu, Y. Sun, and Y.-Y. Liu, Reply to: Deep reinforced learning heuristic tested on spin-glass ground states: The larger picture, Nat. Commun. 14, 5659 (2023).
  27. A. K. Hartmann and A. P. Young, Metastate analysis of the ground states of two-dimensional Ising spin glasses, Phys. Rev. E 108, 024142 (2023).
  28. See Supplemental Material at http://link.aps.org/supplemental/10.1103/kcbw-6zkv for scaling arguments, extended numerical fits, contraction-solver details, multibody correlations, MVC mapping, and additional model studies, which includes Refs. [29–35].
  29. G. H. Wannier, Antiferromagnetism. The triangular Ising net, Phys. Rev. 79, 357 (1950).
  30. J. Villain, Spin glass with non-random interactions, J. Phys. C 10, 1717 (1977).
  31. M. E. Fisher, Statistical mechanics of dimers on a plane lattice, Phys. Rev. 124, 1664 (1961).
  32. M.-S. Vaezi, G. Ortiz, M. Weigel, and Z. Nussinov, The binomial spin glass, Phys. Rev. Lett. 121, 080601 (2018).
  33. A. Lucas, Ising formulations of many NP problems, Front. Phys. 2, 5 (2014).
  34. J. Chen, L. Kou, and X. Cui, An approximation algorithm for the minimum vertex cover problem, Procedia Eng. 137, 180 (2016).
  35. J. Chen, R. Zhou, J. Wu, H. Zhang, and W. Gui, Vertex cover of networks and its related optimization problems: An overview, IEEE Trans. Neural Networks Learn. Syst. 36, 11738 (2024).
  36. H. Khoshbakht and M. Weigel, Domain-wall excitations in the two-dimensional Ising spin glass, Phys. Rev. B 97, 064410 (2018).
  37. V. Kolmogorov, blossom v: A new implementation of a minimum cost perfect matching algorithm, Math. Program. Comput. 1, 43 (2009).
  38. Gurobi Optimization, LLC, Gurobi Optimizer Reference Manual, 2022.

  39. While we will largely focus on the central nearest neighbor bond (with i=i0 and j=j0), in the End Matter we discuss more general situations in which the sites i and j are a distance rij>1 apart. In order to avoid cumbersome notation, in the main text, we will often dispense with “0” subscripts.

  40. For the largest values of Lsub=O(L), the right-hand side of Eq. (2) constitutes an upper bound on the error Eij. Clearly, when Lsub=L (when the subsystem is the entire system), the error Eij=0 by definition.

  41. M. Yannakakis, Equilibria, fixed points, and complexity classes, Comput. Sci. Rev. 3, 71 (2009).
  42. G. Parisi, Infinite number of order parameters for spin-glasses, Phys. Rev. Lett. 43, 1754 (1979).
  43. C. M. Newman and D. L. Stein, Ground state stability and the nature of the spin glass phase, Phys. Rev. E 105, 044132 (2022).

  44. A very similar behavior is observed for the distance |Jc,ij−Jij|, see the discussions in Supplemental Material Sec. S10, but this quantity is not accessible from the perspective of the local solver.

  45. J. P. Sethna, K. A. Dahmen, and C. R. Myers, Crackling noise, Nature (London) 410, 242 (2001).
  46. A. Klaus, S. Yu, and D. Plenz, Statistical analyses support power law distributions found in neuronal avalanches, PLoS One 6, e19779 (2011).
  47. J. M. Carlson and J. Doyle, Highly optimized tolerance: Robustness and design in complex systems, Phys. Rev. Lett. 84, 2529 (2000).
  48. V. Navas-Portella, I. Serra, A. Corral, and E. Vives, Increasing power-law range in avalanche amplitude and energy distributions, Phys. Rev. E 97, 022134 (2018).
  49. L. Fernandez, V. Martin-Mayor, G. Parisi, and B. Seoane, Temperature chaos in 3D Ising spin glasses is driven by rare events, Europhys. Lett. 103, 67003 (2013).
  50. M. Baity-Jesi et al., Temperature chaos is present in off-equilibrium spin-glass dynamics, Commun. Phys. 4, 74 (2021).
  51. H. G. Katzgraber and F. Krzakala, Temperature and disorder chaos in three-dimensional Ising spin glasses, Phys. Rev. Lett. 98, 017201 (2007).
  52. D. Gamarnik, The overlap gap property: A topological barrier to optimizing over random structures, Proc. Natl. Acad. Sci. U.S.A. 118, e2108492118 (2021).
  53. M. Girvan and M. E. J. Newman, Community structure in social and biological networks, Proc. Natl. Acad. Sci. U.S.A. 99, 7821 (2002).
  54. Santo Fortunato, Community detection in graphs, Phys. Rep. 486, 75 (2010).
  55. P. Ronhovde and Z. Nussinov, Multiresolution community detection for mega-scale networks by information-based replica correlations, Phys. Rev. E 80, 016109 (2009).
  56. P. Ronhovde and Z. Nussinov, Local resolution-limit-free model for community detection, Phys. Rev. E 81, 046114 (2010).
  57. T. H. Berlin and M. Kac, The spherical model of a ferromagnet, Phys. Rev. 86, 821 (1952).
  58. J. M. Kosterlitz, D. J. Thouless, and R. C. Jones, Spherical model of a spin glass, Physica B+C (Amsterdam) 86–88, 859 (1977).
  59. A. Harju, B. Barbiellini, S. Siljamäki, R. M. Nieminen, and G. Ortiz, Stochastic gradient approximation: An efficient method to optimize many-body wave functions, Phys. Rev. Lett. 79, 1173 (1997).
  60. S. Ruder, An overview of gradient descent optimization algorithms, 2016, arXiv:1609.04747.
  61. L. Bottou, F. E. Curtis, and J. Nocedal, Optimization methods for large-scale machine learning, SIAM Rev. 60, 223 (2018).
  62. K. Choudhary et al., Recent advances and applications of deep learning methods in materials science, npj Comput. Mater. 8, 59 (2022).
  63. Y. Song, T. Lukasiewicz, Z. Xu, and R. Bogacz, Can the brain do backpropagation? Exact implementation of backpropagation in predictive coding networks, in Advances in Neural Information Processing Systems (NeurIPS), edited by H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, and H. Lin (Curran Associates, New York, 2020).
  64. I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning (MIT Press, Cambridge, MA, 2016).
  65. F. Alet, T. Lozano-Perez, and L. P. Kaelbling, Modular meta-learning, In 2nd Conference on Robot Learning (CoRL), (Zurich, Switzerland, 2018).
  66. D. Meunier, R. Lambiotte, and E. T. Bullmore, Modular and hierarchically modular organization of brain networks, Front. Neurosci. 4, 200 (2010).
  67. D. A. Huse and I. Morgenstern, Finite-size scaling study of the two-dimensional Ising spin glass, Phys. Rev. B 32, 3032 (1985).
  68. A. C. Carter, A. J. Bray, and M. A. Moore, On the use of finite-size scaling to measure spin-glass exponents, J. Phys. A 36, 5699 (2003).
  69. F. Parisen Toldin, A. Pelissetto, and E. Vicari, Finite-size scaling in two-dimensional Ising spin glass models, Phys. Rev. E 84, 051116 (2011).
  70. T. Aspelmeier, H. G. Katzgraber, D. Larson, M. A. Moore, M. Wittmann, and J. Yeo, Finite-size critical scaling in Ising spin glasses in the mean-field regime, Phys. Rev. E 93, 032123 (2016).

  71. We emphasize that the collected data and their fits were for the finite values of |Jc,ijsub−Jij| shown in the figure. When Jc,ijsub≠Jc,ij, the error rate Eij→1/2 as |Jc,ijsub−Jij|→0+, while coincident thresholds Jc,ijsub=Jc,ij (as invariably occurs in the open 1D chain) yield zero error in this limit. At exact GS degeneracy, numerical solvers need not sample both GSs equally. For the finite abscissa values in Fig. 4, the 3D data lies closer to 1/2 than the 2D data.

  72. O. Melchert and A. K. Hartmann, Scaling behavior of domain walls at the zero-temperature ferromagnet to spin-glass transition, Phys. Rev. B 79, 184402 (2009).

Outline

Information

Phys. Rev. Lett. 137, 167401– Published 14 October, 2026
  • Received 1 June 2025
  • Accepted 16 July 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation