Pith. sign in

REVIEW 3 major objections 5 minor 128 references

Ground state energies of multipartite $p$-spin models -- partially lifted RDT view

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read For fully spherical multipartite pure $p$-spin models with even $p$, the ground-state energy is exactly $\sqrt{p}\,u_{\mathrm{GS}}$, where $u_{\mathrm{GS}}$ is the zero of an explicit large-deviation rate function.

desk verdict A real new bounding mechanism and a clean proof that Subag's and Dartois-McKenna's spherical formulas coincide, but the exactness claim outruns what is proved: Theorem 2 is one-sided and the lower bound is imported. read the letter →

arxiv 2509.05916 v1 pith:TJNU3PRY submitted 2025-09-07 math.PR cond-mat.dis-nncs.ITmath-phmath.ITmath.MP

classification math.PRcond-mat.dis-nncs.ITmath-phmath.ITmath.MP MSC 60K3582B4460F10
keywords multipartitep-spinmodelsgroundstateenergysphericalspinglassesIsingrandomdualitytheorylargedeviationsGaussianprocesscomparisontensors
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

This paper proves exact ground-state energies for multipartite pure $p$-spin models when every spin group is constrained to the unit sphere. It constructs two explicit bounds, one from above and one from below, valid for arbitrary spin sets, and shows that on the sphere the two bounds coincide. The resulting value, $\xi_{\mathrm{sph}}(p)=\sqrt{p}\,u_{\mathrm{GS}}$ with $u_{\mathrm{GS}}$ the point where the large-deviation rate function in Eq. (61) crosses zero, is then verified algebraically to equal earlier TAP-based, spectral, and balanced lower-bound predictions. Numerical evidence further suggests that the same two bounds also match for Ising spin sets, which would settle the Ising multipartite ground state up to a single-partite computation. The significance is that a parameter-free sandwich replaces heavy numerical evaluation for multipartite $p$-spin models and, where it closes, gives the exact ground state.

What carries the argument

The load-bearing object is a pair of surrogate Gaussian processes used to bracket the true $p$-tensor process. For the upper bound, the original process is compared with $G_u(\bar{x})=\sum_{j=1}^p (g^{(j)})^\top x^{(j)}$, a sum of independent linear forms; for the lower bound, it is compared with $G_l(\bar{x})=\sum_{j=1}^p \sum_{i_1,\dots,i_p} A^{(j)}_{i_1,\dots,i_p}\prod_{k=1}^p x^{(j)}_{i_k}$, a sum of $p$ independent $p$-spin processes. The comparison rests on the elementary inequality $\prod_{j=1}^p a_j + p - 1 - \sum_{j=1}^p a_j \ge 0$ for $a_j\in[-1,1]$, which implies that the true process dominates $G_l$ and is dominated by $G_u$ in the Gaussian comparison sense. Optimizing the auxiliary parameter $c_3$ converts these comparisons into explicit upper and lower bounds on the ground-state energy. In the spherical case the bounds are evaluated through a Gaussian-norm large-deviation estimate, producing the rate function $\varphi_{\mathbb{S}^n}(p,u)$; setting this rate function to zero defines $u_{\mathrm{GS}}$ and yields the exact value. The same machinery gives candidate Ising bounds via a complementary-error-function estimate.

What would settle it

For $p=4$ Ising spins, compute the true multipartite ground-state energy independently (for instance by a direct numerical optimization of the Parisi-type variational problem for the four-part model or by finite-size extrapolation) and compare it with the listed candidate $\xi^{(2,p)}_{\mathrm{sk}}(4)=2.3348$; an exact value strictly below that number would disprove the claim that the Ising bounds match.

Watch

Extended reading notes

Core claim

The paper's central claim is that the ground-state energy of a multipartite pure $p$-spin model, with $p$ interacting spin vectors each on the unit sphere, is exactly $\xi_{\mathrm{sph}}(p)=\sqrt{p}\,u_{\mathrm{GS}}$, where $u_{\mathrm{GS}}$ is the smallest $u$ such that the rate function $\varphi_{\mathbb{S}^n}(p,u)$ from Eq. (61) is negative. The equality is obtained by sandwiching the true energy between a lower bound built from $p$ independent $p$-spin processes and an upper bound built from $p$ independent linear processes; both bounds are derived from Gaussian covariance comparisons and then optimized. On the sphere the two optimized bounds meet at the same number. The paper then shows by direct algebra that this common value coincides with the TAP-equation prediction [116], the tensor upper bound [39], and the balanced multi-species lower bound [26], so the earlier numerical coincidence between [116] and [39] becomes an analytic identity.

Load-bearing premise

The spherical exactness is imported at the lower end: it assumes the earlier critical-point-complexity and balanced multi-species lower bounds apply to this multipartite pure p-spin setting with the same normalization; without that assumption the paper establishes only an upper bound, not the exact value.

Editorial extensions

If this is right

  • For arbitrary spin sets, the multipartite ground-state energy is bracketed by two closed-form bounds that avoid the exponential numerical cost of the fully lifted formulation.
  • For fully spherical sets, the bounds coincide and give exact values $\xi_{\mathrm{sph}}(p)=\sqrt{p}\,u_{\mathrm{GS}}$ for even $p$; concrete values are listed for $p=2,\dots,7$.
  • The spherical value agrees analytically with the TAP prediction [116], the upper bound [39], and the lower bound [26], so the previously numerical agreement between [116] and [39] is now a proven equality.
  • Whenever a spin set's single-partite ground state is attained at the second partial lifting level, the multipartite ground state equals $\sqrt{p}$ times the single-partite one (Corollary 3).
  • Numerical evidence suggests the Ising bounds also match, so the Ising multipartite ground state would be $\sqrt{p}\,u_{\mathrm{GS}}^{\mathrm{(sk)}}$ provided the single-partite bound in Eq. (101) is tight.

Reading between the lines

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

  • One can test the same bound pair on intermediate spin sets, such as the intersection of a scaled cube $\{x:x_i^2\le c/n\}$ with the unit sphere; matching there would delineate how far the exactness extends beyond spherical and Ising sets.
  • The $\sqrt{p}$ factor is likely a universal feature: whenever the multipartite disorder decomposes into $p$ independent single-partite copies, the ground-state energy should be $\sqrt{p}$ times the single-partite value, independent of the spin set.
  • If the Ising numerical agreement survives scrutiny, the paper's second-level-lifting criterion could become a practical algorithm: compute a single-partite ground state and multiply by $\sqrt{p}$, instead of solving the full multipartite problem.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes a 'partially lifted random duality theory' (pl RDT) mechanism to produce upper and lower bounds for ground state energies of multipartite pure p-spin models with arbitrary spin sets. A Gaussian comparison argument (Theorem 1) gives non-asymptotic bounds, and specializing to identical spherical or Ising spin sets yields large deviation upper bounds for the normalized maximum of the p-spin Hamiltonian (Theorems 2 and 3). The paper claims that for fully spherical sets the upper and lower bounds match, giving exact ground state energies that agree with Subag's TAP prediction, the Dartois-McKenna upper bound, and the Bates-Sohn lower bound; for Ising sets it presents numerical evidence and an explicitly conditional equality.

Significance. If the exactness claim for spherical multipartite pure p-spin models were fully proved, the paper would provide a clean, unified derivation of known ground state energy formulas via a relatively simple large deviation mechanism. The upper-bound side and the algebraic matching with [39] and [116] are valuable and largely self-contained. The paper is also honest in stating the Ising equality as conditional and in posing open questions about set structures for which the bounds match. However, the central spherical exactness claim is not established within the manuscript: only a one-sided large deviation bound is proved, and the matching lower bound is imported from external results without the required normalization-specific verification.

major comments (3)
  1. [Section 3.1, Eqs. (43), (61)-(65)] Theorem 2 proves only the one-sided bound limsup_n (1/n) log P(zeta(p; S_n, n) >= u) <= phi_Sn(p,u). The subsequent identities xi_sph(p) = u_GS and xi_sph(p) = sqrt(p) u_GS require the matching lower direction, i.e. that zeta(p; S_n, n) is at least u_GS - epsilon with probability tending to 1 at the normalization of Eq. (33). That lower bound is not derived. The appeal to the expected critical-point complexity of [9] and to the balanced multi-species lower bound of [26] supplies, at best, annealed information or results for different normalizations; no second-moment or quenched argument is given for the identical-sphere multipartite normalization. Since the abstract's claim that the bounds 'actually match' in the spherical case rests on this missing direction, this is a load-bearing gap.
  2. [Section 3.2, Eqs. (99)-(102)] The Ising equality is explicitly conditional: the text states 'Provided that (101) holds with equality', and no proof of this equality is supplied. While the paper correctly labels the Ising matching as numerical evidence, the same missing lower-bound mechanism underscores that the paper, as written, establishes upper bounds rather than exact ground state energies for the Ising specialization. This should be clearly separated from the proved results in the abstract and conclusion.
  3. [Section 3.3, Corollary 3, Eq. (103)] Corollary 3 characterizes the matching condition as 'GSE(pSP(S)) is achieved on the second partial level of lifting'. In the spherical case this condition is asserted to hold on the basis of the agreement with the critical-point exponent of [9], but as noted above the lower-bound direction is not proved for the normalization of Eq. (33). Thus the universality conclusion and the two 'interesting questions' in Section 3.3 are built on an unverified condition; the corollary is conditional rather than a demonstrated result.
minor comments (5)
  1. [Theorem 1, statement and Eq. (6)] The statement says A^(j) in R^(n x n), but in Eq. (6) and Eq. (18) A^(j) is indexed by p indices and used as an n^p tensor; this should be corrected to R^(n^p) or otherwise clarified.
  2. [Eq. (13)] After taking k = p-1, the factor (1 - a^(j)) contains an undefined index j; it should be (1 - a^(p-1)) or the appropriate index should be named.
  3. [Section 3.2, Eq. (90)] The definition of hat c_3 as an argmin includes the term -c_3 u, but the variable u is not introduced in the preceding text; it should be stated that this is for a fixed u in the large deviation bound.
  4. [Section 3.3, final paragraph] There is a typo in 'ovperall role' which should read 'overall role'.
  5. [Figures and Tables] Figure 2 and Table 2 would benefit from a statement that the displayed Ising values are upper bounds or conditional values, since the surrounding text sometimes refers to them simply as GSE values.

Circularity Check

0 steps flagged · score 2.0 of 10

No definitional circularity: the spherical exact value is imported from external lower-bound results, and the only self-citation is an auxiliary large-deviation evaluation.

full rationale

The derivation chain is not circular in the sense of reducing a prediction to its own inputs. Theorem 1 obtains upper and lower bounding functionals through Gordon comparisons and elementary algebra, with the lower and upper functionals being genuinely different Gaussian processes rather than rescalings of the target GSE. Theorem 2 proves only a Chernoff/Markov limsup upper bound on the normalized tail probability; it does not assert the matching lower LDP inside that theorem. The subsequent identification of the exact spherical GSE in Eqs. (62)-(65) explicitly leans on the connection to Auffinger-Ben Arous-Cerny and the external lower bounds of Bates-Sohn, and the agreement with Subag and Dartois-McKenna is verified by explicit algebra. The only self-citation that enters the proof is the Gaussian norm large-deviation evaluation at Eq. (46), attributed to the author's own works [105,107]; this is a parameter-free auxiliary calculation, independently checkable, and is not the target GSE formula. The Ising specialization is explicitly conditional ('Provided that (101) holds with equality'), so no overclaim is disguised. The main weakness is that the lower bound for exactness is imported from external results rather than proved in the paper, but that is a rigor/completeness concern rather than circularity. Score 2 reflects the minor self-citation in the LDP derivation without treating it as load-bearing circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No free parameters or invented entities. The derivation introduces variational parameters (c3, γ) that are optimized, not fitted. The exactness claim for spherical spins borrows its lower bound from cited external results, and the Ising exactness is left conditional.

assumptions (3)
  • standard math Gaussian comparison inequalities (Slepian/Gordon type) allow comparing maxima of centered Gaussian processes when covariances are ordered (used in Theorem 1, Eqs. (14) and (22)).
    Invoked via [53]; this is the engine of the upper and lower bound derivation.
  • domain assumption The spherical multipartite pure p-spin GSE is correctly determined by the lower bound of Bates-Sohn [26] and the critical point complexity characterization of Auffinger-Ben Arous-Cerny [9].
    Used to conclude equality in the spherical case; without these, Theorem 2 only supplies an upper bound.
  • domain assumption For Ising spins the Parisi formula gives the exact GSE, so an upper bound matching it would establish equality.
    Invoked in Section 3.2 to suggest tightness of the upper bound via numerical Parisi optimization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ground state energies of multipartite $p$-spin models -- partially lifted RDT view." pith.science (2026). https://pith.science/paper/TJNU3PRY

@misc{pith2026250905916,
  author       = {Pith},
  title        = {Pith review of: Ground state energies of multipartite $p$-spin models -- partially lifted RDT view},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TJNU3PRY}},
  note         = {Machine review of arXiv:2509.05916}
}
abstract

We consider ground state energies (GSE) of multipartite $p$-spin models. Relying on partially lifted random duality theory (pl RDT) concepts we introduce an analytical mechanism that produces easy to compute lower and upper GSE bounds for \emph{any} spin sets. We uncover that these bounds actually match in case of fully spherical sets thereby providing optimal GSE values for spherical multipartite pure $p$-spin models. Numerical evidence further suggests that our upper and lower bounds may match even in the Ising scenarios. As such developments are rather intriguing, we formulate several questions regarding the connection between our bounds matching generality on the one side and the spin sets structures on the other.

Figures

Figures reproduced from arXiv: 2509.05916 by the authors.

Figure 1
Figure 1. ξsph(p) as a function of p 14 [PITH_FULL_IMAGE:figures/full_fig_p014_1.png] view at source ↗
Figure 2
Figure 2. ξsk(p) (bound) as a function of p are needed. Things do simplify a bit when S (j) = S, 1 ≤ j ≤ p, but the need for quicker more practical characterization remains. The bounds that we presented above are particularly useful in such contexts as they are substantially simpler. To have them fully operational, an assessment of their accuracy is welcome as well. The fact that there are scenarios where they provide the exa… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

128 extracted references · 43 canonical work pages

  1. [39]

    Dartois and B

    S. Dartois and B. McKenna. Injective norm of real and com plex random tensors I: From spin glasses to geometric entanglement. 2024. available online at http://arxiv.org/abs/2404.03627

  2. [116]

    E. Subag. TAP approach for multispecies spherical spi n glasses II: The free energy of the pure models. The Annals of Probability , 51(3):1004 – 1024, 2023

  3. [9]

    Auffinger, G

    A. Auffinger, G. Ben Arous, and J. Cerny. Random matrices an d complexity of spin glasses. Commu- nications on Pure and Applied Mathematics , 66(2):165–201, 2013

  4. [26]

    Bates and Y

    E. Bates and Y. Sohn. Balanced multi-species spin glass es. 2025. available online at http://arxiv. org/abs/2507.06522

  5. [1]

    Aizenman, R

    M. Aizenman, R. Sims, and S. L. Starr. Extended variation al principle for the Sherrington-Kirkpatrick spin-glass model. Phys. Rev. B , 68:214403, Dec 2003

  6. [2]

    El Alaoui, A

    A. El Alaoui, A. Montanari, and M. Sellke. Sampling from t he Sherrington-Kirkpatrick gibbs measure via algorithmic stochastic localization. In 63rd IEEE Annual Symposium on Foundations of Computer Science, FOCS 2022, Denver, CO, USA, October 31 - November 3, 2 022, pages 323–334. IEEE, 2022

  7. [3]

    El Alaoui, A

    A. El Alaoui, A. Montanari, and M. Sellke. Shattering in p ure spherical spin glasses. Communications in Mathematical Physics , 406(111), 2025

  8. [4]

    Alberici, F

    D. Alberici, F. Camilli, P. Contucci, and E. Mingione. Th e multi-species mean-field spin-glass on the Nishimori line. Journal of Statistical Physics , 182, 01 2021

Show all 128 references
  1. [5]

    S. Amari. Learning patterns and pattern sequences by sel f-organizing nets of threshold elements. IEEE Transactions on computers , 100:1197 – 1206, 1972

  2. [6]

    D. Amit, H. Gutfreund, and H. Sompolinsky. Statistical m echanics of neural networks. Annals of Physics, 173:30–47, 1987

  3. [7]

    D. J. Amit, H. Gutfreund, and H. Sompolinsky. Storing infi nite number of patterns in a spin glass model of neural networks. Phys. Rev. Letters , 55:1530, 1985

  4. [8]

    Ben Arous, S

    G. Ben Arous, S. Mei, A. Montanari, and M. Nica. The landsc ape of the spiked tensor model. Com- munications on Pure and Applied Mathematics , 72:2282– 2330, 2019

  5. [10]

    Auffinger and G

    A. Auffinger and G. Ben Arous. Complexity of random smooth functions on the high-dimensional sphere. The Annals of Probability , 41:4214 – 4247, 2013

  6. [11]

    Auffinger and W.-K

    A. Auffinger and W.-K. Chen. Free energy and complexity of spherical bipartite models. J. Stat. Phys. , 157:40 – 59, 2014

  7. [12]

    Auffinger and W.-K

    A. Auffinger and W.-K. Chen. The Parisi formula has a uniqu e minimizer. Communications in Mathematical Physics, 335(3), 2015

  8. [13]

    Auffinger and W.-K

    A. Auffinger and W.-K. Chen. Parisi formula for the ground state energy in the mixed p-spin model. Ann. Probab., 45:4617 – 4631, 2017

  9. [14]

    Auffinger, W.-K

    A. Auffinger, W.-K. Chen, and Q. Zeng. The SK model is infini te step replica symmetry breaking at zero temperature. Comm. Pure Appl. Math. , 73:921 – 943, 2020

  10. [15]

    Baik and J

    J. Baik and J. O. Lee. Free energy of bipartite spherical Sherrington-Kirkpatrick model. Ann. Inst. H. Poincaré Probab. Statist. , 56:2897 – 2934, 2020. 18

  11. [16]

    A. S. Bandeira, S. Gopi, H. Jiang, and K. Lucca. A geometr ic perspective on the injective norm of sums of random tensors. 2024. available online at http://arxiv.org/abs/2411.10633

  12. [17]

    Barnum and N

    H. Barnum and N. Linden. Monotones and invariants for mu lti-particle quantum states. Physical Review A , 34:6787, 2001

  13. [18]

    Barra, P

    A. Barra, P. Contucci, E. Mingione, and D. Tantari. Mult i-Species Mean Field Spin Glasses. Rigorous Results. Annales Henri Poincaré , 16:691 – 708, 3 2015

  14. [19]

    Barra, G

    A. Barra, G. Genovese, and F. Guerra. The replica symmet ric approximation of the analogical neural network. J. Stat. Physics , 140(4):784–796, 2010

  15. [20]

    Barra, G

    A. Barra, G. Genovese, and F. Guerra. Equilibrium stati stical mechanics of bipartite spin systems. Journal of Physics A: Mathematical and Theoretical , 44(24):245002, 2011

  16. [21]

    Barra, G

    A. Barra, G. Genovese, F. Guerra, and D. Tantari. How gla ssy are neural networks? Journal of Statistical Mechanics: Theory and Experiment , 2012(07):P07009, 2012

  17. [22]

    Barra, G

    A. Barra, G. Genovese, P. Sollich, and D. Tantari. Phase diagram of restricted Boltzmann machines and generalized hopfield networks with arbitrary priors. Phys. Rev. E , 97:022310, 2018

  18. [23]

    Bates, L

    E. Bates, L. Sloman, and Y. Sohn. Replica symmetry break ing in multi-species Sherrington-Kirkpatrick model. J. Stat. Phys. , 2:333 – 350, 2019

  19. [24]

    Bates and Y

    E. Bates and Y. Sohn. Crisanti-Sommers formula and simu ltaneous symmetry breaking in multi-species spherical spin glasses. Communications in Mathematical Physics , 394:1101 – 1152, 3 2022

  20. [25]

    Bates and Y

    E. Bates and Y. Sohn. Free energy in multi-species mixed p-spin spherical models. Electron. J. Probab., 27:1 – 75, 2022

  21. [27]

    Bovier and V

    A. Bovier and V. Gayrard. Hopfield models as generalized random mean field models. In mathematical aspects of spin glasses and neural networks, Progr. Prob. , 41:3–89, 1998

  22. [28]

    Brunetti, G

    R. Brunetti, G. Parisi, and F. Ritort. Asymmetric Littl e spin glas model. Physical Review B , 46(9), September 1992

  23. [29]

    H.-B. Chen, V. Issa, and J.-C. Mourrat. The convex struc ture of the Parisi formula for multi-species spin glasses. 2025. available online at http://arxiv.org/abs/2508.06397

  24. [30]

    Chen and J.-C

    H.-B. Chen and J.-C. Mourrat. On the free energy of vecto r spin glasses with nonconvex interactions. Probability and Mathematical Physics , 6(1):1 – 80, 2025

  25. [31]

    W.-K. Chen. The Aizenman-Sims-Starr scheme and Parisi formula for mixed p-spin spherical models. Electronic Journal of Probability , 18:1 – 14, 2013

  26. [32]

    Chen and Sen A

    W.-K. Chen and Sen A. Parisi formula, disorder chaos and fluctuation for the ground state energy in the spherical mixed p-spin models. Commun. Math. Phys. , 350:129 – 173, 2017

  27. [33]

    W.-K. Chen, M. Handschy, and G. Lerman. Phase transitio n in random tensors with multiple inde- pendent spikes. The Annals of Applied Probability , 31(4):1868–1913, 2021

  28. [34]

    Chen and D

    W.-K. Chen and D. Panchenko. On the TAP Free Energy in the Mixed p-Spin Models. Communications in Mathematical Physics , 362:219 – 252, 1 2018

  29. [35]

    W.-K. Chen, D. Panchenko, and E. Subag. The generalized tap free energy II. Communications in Mathematical Physics, 381:1–35, 01 2021

  30. [36]

    W.-K. Chen, D. Panchenko, and E. Subag. Generalized tap free energy. Communications on Pure and Applied Mathematics, 76(7):1329–1415, 2023. 19

  31. [37]

    Contucci and I

    P. Contucci and I. Gallo. Bipartite mean field spin syste ms. existence and solution. Mathematical Physics Electronic Journal, , 14, 1 2008

  32. [38]

    Crisanti and H

    A. Crisanti and H. J. Sommers. The spherical p-spin inte raction spin glass model: the statics. Zeitschrift für Physik B Condensed Matter , 87:341–354, 1992

  33. [40]

    Dean and F

    D. Dean and F. Ritort. Squared interaction matrix Sherr ington-Kirkpatrick model for a spin glass. Phys. Rev. B , 65:224209, 2002

  34. [41]

    B. Derrida. Random-energy model: Limit of a family of di sordered models. Physical Review Letters 45, 45:79, 1980

  35. [42]

    B. Derrida. Random-energy model: An exactly solvable m odel of disordered systems. Physical Review B 2 , 24:2613, 1981

  36. [43]

    Dey and Q

    P. Dey and Q. Wu. Fluctuation results for multi-species Sherrington-Kirkpatrick model in the replica symmetric regime. Journal of Statistical Physics , 185, 12 2021

  37. [44]

    S. F. Edwards and P. W. Anderson. Theory of spin glasses. Journal of Physics F: Metal Physics , 5(5):965, may 1975

  38. [45]

    Fedele and P

    M. Fedele and P. Contucci. Scaling limits for multi-spe cies statistical mechanics mean-field models. Journal of Statistical Physics , 144:1186–1205, 6 2011

  39. [46]

    Fedele and F

    M. Fedele and F. Unguendoli. Rigorous results on the bip artite mean-field model. Journal of Physics A: Mathematical and Theoretical , 45(38):385001, 2012

  40. [47]

    J. Feng, M. Shcherbina, and B. Tirozzi. On the critical c apacity of the Hopfield model. Communications in Mathematical Physics , 2000

  41. [48]

    Fyodorov

    Yan V. Fyodorov. Complexity of random energy landscape s, glass transition, and absolute value of the spectral determinant of random matrices. Phys. Rev. Lett. , 92:2400601, 2004

  42. [49]

    Gamarnik

    D. Gamarnik. The overlap gap property: A topological ba rrier to optimizing over random structures. Proceedings of the National Academy of Sciences , 118(41), 2021

  43. [50]

    Gamarnik and A

    D. Gamarnik and A. Jagannath. The overlap gap property a nd approximate message passing algorithms for p-spin models. Ann. Probab., 49:180 – 205, 2021

  44. [51]

    Gamarnik, A

    D. Gamarnik, A. Jagannath, and E. C. Kızıldag. Shatteri ng in the Ising pure p-spin model. 2024. available online at http://arxiv.org/abs/2307.07461

  45. [52]

    Genovese

    G. Genovese. A remark on the spherical bipartite spin gl ass. Mathematical Physics, Analysis and Geometry, 25:14, 2022

  46. [53]

    Y. Gordon. Some inequalities for Gaussian processes an d applications. Israel Journal of Mathematics , 50(4):265–289, 1985

  47. [54]

    Grothendieck

    A. Grothendieck. Résumé des résultats essentiels dans la théorie des produits tensoriels topologiques et des espaces nucléaires. Annales de l’Institut Fourier , 4:73 – 112, 1952

  48. [55]

    F. Guerra. Broken replica symmetry bounds in the mean fie ld spin glass model. Comm. Math. Physics , 233:1–12, 2003

  49. [56]

    D. O. Hebb. Organization of behavior. New York: Wiley , 1949

  50. [57]

    J. J. Hopfield. Neural networks and physical systems wit h emergent collective computational abilities. Proc. Nat. Acad. Science , 79:2554, 1982. 20

  51. [58]

    Huang and M

    B. Huang and M. Sellke. Tight lipschitz hardness for opt imizing mean field spin glasses. In 63rd IEEE Annual Symposium on Foundations of Computer Science, FOCS 2 022, Denver, CO, USA, October 31 - November 3, 2022 , pages 312–322. IEEE, 2022

  52. [59]

    Huang and M

    B. Huang and M. Sellke. Algorithmic threshold for multi -species spherical spin glasses. 2023. available online at http://arxiv.org/abs/2303.12172

  53. [60]

    Huang and M

    B. Huang and M. Sellke. A constructive proof of the spher ical Parisi formula. 2023. available online at http://arxiv.org/abs/2311.15495

  54. [61]

    Huang and M

    B. Huang and M. Sellke. Optimization algorithms for mul ti-species spherical spin glasses. J. Stat. Phys., 191, 2024

  55. [62]

    V. Issa. Existence and uniqueness of permutation-inva riant optimizers for Parisi formula. 2024. avail- able online at http://arxiv.org/abs/2407.13846

  56. [63]

    Jagannath, P

    A. Jagannath, P. Lopatto, and L. Miolane. Statistical t hresholds for tensor pca. The Annals of Applied Probability, 30(4):1910–1933, 2020

  57. [64]

    Jagannath and I

    A. Jagannath and I. Tobasco. A dynamic programming appr oach to the Parisi functional. In Proceed- ings of the American Mathematical Society , volume 14, pages 3135–3150

  58. [65]

    Jagannath and I

    A. Jagannath and I. Tobasco. Low temperature asymptoti cs of spherical mean field spin glasses. Communications in Mathematical Physics , 352:979 – 1012, 3 2017

  59. [66]

    Kirkpatrick and D

    S. Kirkpatrick and D. Sherrington. Infinite-ranged mod els of spin-glasses. Phys. Rev. B , 17:4384–4403, Jun 1978

  60. [67]

    P. Kivimae. The ground state energy and concentration o f complexity in spherical bipartite models. Communications in Mathematical Physics , 403:37 – 81, 1 2023

  61. [68]

    J. Ko. Free energy of multiple systems of spherical spin glasses with constrained overlaps. Electronic Journal of Probability , 25:1 – 34, 2020

  62. [69]

    I. Y. Korenblit, Y. A. Fyodorov, and E. F. Shender. Antif erromagnetic spin glass in the Ising model. J. Phys. C: Solid State Phys. , 20:1835 – 1839, 1987

  63. [70]

    I. Y. Korenblit, Y. A. Fyodorov, and E. F. Shender. Phase transitions in frustrated metamagnets. Europhysics Letters (EPL) , 4:827 – 832, 1987

  64. [71]

    I. Y. Korenblit and E. F. Shender. Spin glass in an lsing t wo-sublattice magnet. Zh. Eksp. Teor. Fiz. , 89:1785 – 1795, 1985

  65. [72]

    J. M. Kosterlitz, D. J. Thouless, and Raymund C. Jones. S pherical model of a spin-glass. Phys. Rev. Lett., 36:1217–1220, May 1976

  66. [73]

    Krotov and J

    D. Krotov and J. J. Hopfield. Dense associative memory fo r pattern recognition. Advances in Neural Information Processing Systems , 1:1180–1188, 2016

  67. [74]

    Lesieur, L

    T. Lesieur, L. Miolane, M. Lelarge, F. Krzakala, and L. Z deborová. Statistical and computational phase transitions in spiked tensor estimation. In 2017 IEEE International Symposium on Information Theory (ISIT) , pages 511–515, 2017

  68. [75]

    L.-H. Lim. Tensors in computations. Acta Numerica, 30:555 – 764, 2021

  69. [76]

    W. A. Little. The existence of persistent states in the b rain. Math. Biosci. , 19(1-2):101–120, 1974

  70. [77]

    Loukianova

    D. Loukianova. Capacite de memoire dans le modele de Hop field. C. R. Acad. Sci. Paris t. 318, Serie I, pages 157–160, 1994

  71. [78]

    Loukianova

    D. Loukianova. Lower bounds on the restitution error in the Hopfield model. Probab. Theory Related Fields, 107:161–176, 1997. 21

  72. [79]

    R. J. MacEliece, E. C. Posner, E. Rodemich, and S.S. Venk atesh. The capacity of the Hopfield associative memory. IEEE Trans. Inform. Theory , 33:461–482, 1987

  73. [80]

    B. McKenna. Complexity of bipartite spherical spin gla sses. Annales de l’Institut Henri Poincaré, Probabilités et Statistiques , 60(1):636 – 657, 2024

  74. [81]

    Montanari

    A. Montanari. Optimization of the Sherrington-Kirkpa trick hamiltonian. In 60th IEEE Annual Sympo- sium on Foundations of Computer Science, FOCS 2019, Baltimo re, Maryland, USA, November 9-12, 2019, pages 1417–1433. IEEE Computer Society, 2019

  75. [82]

    J.-C. Mourrat. Free energy upper bound for mean-field ve ctor spin glasses. Annales de l’Institut Henri Poincaré, Probabilités et Statistiques , 59(3):1143 – 1182, 2023

  76. [83]

    J.-C. Mourrat. An informal introduction to the Parisi f ormula. 2024. available online at http:// arxiv.org/abs/2410.12364

  77. [84]

    Mourrat and D

    J.-C. Mourrat and D. Panchenko. Extending the Parisi fo rmula along a Hamilton-Jacobi equation. Electron. J. Probab., 25:1 – 17, 2020

  78. [85]

    Ch. M. Newman. Memory capacity and neural network model s: Rigorous lower bounds. Neural Networks, 1:223–238, 1988

  79. [86]

    Panchenko

    D. Panchenko. A connection between the Ghirlanda-Guer ra identities and ultrametricity. The Annals of Probability, 38(1):327–347, 2010

  80. [87]

    Panchenko

    D. Panchenko. The Ghirlanda-Guerra identities for mix ed p-spin model. Comptes Rendus Mathema- tique, 348(3-4):189–192, 2010

  81. [88]

    Panchenko

    D. Panchenko. The Parisi ultrametricity conjecture. Ann. Math. , 77(1):383–393, 2013

  82. [89]

    Panchenko

    D. Panchenko. The Sherrington-Kirkpatrick model . Springer Science & Business Media, 2013

  83. [90]

    Panchenko

    D. Panchenko. Spin glass models from the point of view of spin distributions. The Annals of Probability, 41(3A):1315–1361, 2013

  84. [91]

    Panchenko

    D. Panchenko. The free energy in a multi-species Sherri ngton-Kirkpatrick model. The Annals of Probability, 43:3494 – 3513, 2015

  85. [92]

    Panchenko

    D. Panchenko. Free energy in the mixed p-spin models wit h vector spins. The Annals of Probability , 46(2):865–896, 2018

  86. [93]

    Panchenko

    D. Panchenko. Free energy in the Potts spin glass. The Annals of Probability , 46:829 – 864, 2018

  87. [94]

    G. Parisi. Infnite number of order parameters for spin- glasses. Phys. Rev. Lett. , 43:1754–1756, 1979

  88. [95]

    G. Parisi. Breaking the symmetry in SK model. J. Physics , A13:1101, 1980

  89. [96]

    G. Parisi. A sequence of approximated solutions to the S K model for spin glasses. Journal of Physics A: Mathematical and General , 13(4):L115, 1980

  90. [97]

    G. Parisi. Order parameter for spin glasses. Phys. Rev. Lett. , 50:1946, 1983

  91. [98]

    Pastur and A

    L. Pastur and A. Figotin. On the theory of disordered spi n systems. Theory Math. Phys. , 35(403-414), 1978

  92. [99]

    Pastur, M

    L. Pastur, M. Shcherbina, and B. Tirozzi. The replica-s ymmetric solution without the replica trick for the Hopfield model. Journal of Statistical Physics , 74(5/6), 1994

  93. [100]

    Ramsauer, B

    H. Ramsauer, B. Schafl, J. Lehner, P. Seidl, M. Widrich, L. Gruber, M. Holzleitner, T. Adler, D. Kreil, M. K. Kopp, G. Klambauer, J. Brandstetter, and S. Hochreiter . Hopfield networks is all you need. In International Conference on Learning Representations , 2021. 22

  94. [101]

    V. Ros, G. Ben Arous, G. Biroli, and C. Cammarota. Compl ex energy landscapes in spiked-tensor and simple glassy models: Ruggedness, arrangements of loca l minima, and phase transitions. Physical Review X , 9:011003, 2019

  95. [102]

    Shcherbina and B

    M. Shcherbina and B. Tirozzi. The free energy of a class of Hopfield models. Journal of Statistical Physics, 72(1/2), 1993

  96. [103]

    Sherrington and S

    D. Sherrington and S. Kirkpatrick. Solvable model of a spin-glass. Phys. Rev. Lett. , 35:1792–1796, Dec 1975

  97. [104]

    A. Shimony. Degree of entanglement. Annals of the New York Academy of Sciences , 755:675 – 679, 1995

  98. [105]

    M. Stojnic. Random linear systems with sparse solutio ns – asymptotics and large deviations. available online at http://arxiv.org/abs/1612.06361

  99. [106]

    M. Stojnic. Asymmetric Little model and its ground sta te energies. 2013. available online at http:// arxiv.org/abs/1306.3978

  100. [107]

    M. Stojnic. Lifting/lowering Hopfield models ground s tate energies. 2013. available online at http:// arxiv.org/abs/1306.3975

  101. [108]

    M. Stojnic. Fully bilinear generic and lifted random p rocesses comparisons. 2016. available online at http://arxiv.org/abs/1612.08516

  102. [109]

    M. Stojnic. Generic and lifted probabilistic compari sons – max replaces minmax. 2016. available online at http://arxiv.org/abs/1612.08506

  103. [110]

    M. Stojnic. Fully lifted interpolating comparisons o f bilinearly indexed random processes. 2023. available online at http://arxiv.org/abs/2311.18092

  104. [111]

    M. Stojnic. Fully lifted random duality theory. 2023. available online at http://arxiv.org/abs/ 2312.00070

  105. [112]

    M. Stojnic. Capacity of the Hebbian-Hopfield network a ssociative memory. 2024. available online at http://arxiv.org/abs/2403.01907

  106. [113]

    The complexity of spherical p-spin models - A s econd moment approach

    E Subag. The complexity of spherical p-spin models - A s econd moment approach. Ann. Probab., 45:3385 – 3450, 2017

  107. [114]

    The geometry of the gibbs measure of pure spher ical spin glasses

    E Subag. The geometry of the gibbs measure of pure spher ical spin glasses. Inventiones Mathematicae, 210:135 – 209, 2017

  108. [115]

    Following the ground states of full-rsb spher ical spin glasses

    E Subag. Following the ground states of full-rsb spher ical spin glasses. Comm. Pure Appl. Math. , 74:1021–1044, 2021

  109. [117]

    The free energy of spherical pure p-spin model s: computation from the TAP approach

    E Subag. The free energy of spherical pure p-spin model s: computation from the TAP approach. Probability Theory and Related Fields , 186:715–734, 2023

  110. [118]

    Free energy landscapes in spherical spin glas ses

    E Subag. Free energy landscapes in spherical spin glas ses. Duke Math. J. , 173:1291 – 1357, 2024

  111. [119]

    E. Subag. TAP approach for multi-species spherical sp in glasses I: General theory. Electronic Journal of Probability, 30:1 – 32, 2025

  112. [120]

    Subag and O

    E. Subag and O. Zeitouni. The extremal process of criti cal points of the pure p-spin spherical spin glass model. Probability Theory and Related Fields , 168(3):773–820, 2017

  113. [121]

    Subag and O

    E. Subag and O. Zeitouni. Concentration of the complex ity of spherical pure p-spin models at arbitrary energies. Journal of Mathematical Physics , 62(12):123301, 12 2021. 23

  114. [122]

    Talagrand

    M. Talagrand. Rigorous results for the Hopfield models with many patterns. Prob. Theor. Rel. Fields , 110:109–176, 1998

  115. [123]

    On Guerra’s broken replica-symmetry bou nd

    M Talagrand. On Guerra’s broken replica-symmetry bou nd. C. R. Math. Acad. Sci. Paris , 337:477 – 480, 2003

  116. [124]

    Talagrand

    M. Talagrand. Free energy of the spherical mean field mo del. Probability Theory and Related Fields , 134:339–382, 3 2006

  117. [125]

    Talagrand

    M. Talagrand. The Parisi formula. Annals of mathematics , 163:221–263, 01 2006

  118. [126]

    Talagrand

    M. Talagrand. Mean field models and spin glasse: Volume II . A series of modern surveys in mathematics 55, Springer-Verlag, Berlin Heidelberg, 2011

  119. [127]

    Talagrand

    M. Talagrand. Mean field models and spin glasses: Volume I . A series of modern surveys in mathematics 54, Springer-Verlag, Berlin Heidelberg, 2011

  120. [128]

    Wei and P

    T.-C. Wei and P. M. Goldbart. Geometric measure of enta nglement and applications to bipartite and multipartite quantum states. Physical Review A , 68:042307, 2003. 24

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.