Pith. sign in

REVIEW 5 minor 47 references

Continuity and Discontinuity of McKean-Vlasov Phase Transitions via Bifurcation Theory

T0 review · 0 major / 5 minor · reviewed 2026-07-14 · grok-4.5

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

Pith's one-line read Continuous McKean–Vlasov phase transitions are pitchforks; discontinuous ones are saddle-nodes, classified by a covariance-eigenvalue criterion.

desk verdict Solid local bifurcation classification of continuous vs discontinuous McKean–Vlasov transitions in R^n, with a clean nD Dawson criterion; scope is deliberately 1D-null-space and the Gaussian case is numerical only. read the letter →

arxiv 2607.10723 v1 pith:B4NAFQ2B submitted 2026-07-12 math.DS math.APmath.PR

Junlang Hu , Zhenxin Liu This is my paper · ORCID
classification math.DSmath.APmath.PR MSC 35Q8335Q7034K1882C22
keywords McKean-VlasovSDEphasetransitionspitchforkbifurcationsaddle-nodeDawson'scriterioninvariantmeasuresnonlocalFokker-Planck
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 classifies how the invariant measures of McKean–Vlasov dynamics in Euclidean space change with temperature when a non-convex confining potential competes with cooperative interactions. Continuous transitions, in which new measures peel smoothly off a trivial branch, are shown to be pitchfork bifurcations; discontinuous transitions, in which a measure appears abruptly in a shallower well, are saddle-node bifurcations. The authors extend Dawson’s classical one-dimensional criticality condition to n dimensions: the critical temperature is fixed by a simple eigenvalue of the covariance matrix of the critical measure, and the direction of the bifurcation is the corresponding eigenvector. Concrete double-well, four-well and Gaussian-interaction examples are recovered from the abstract theorems, so the same spectral test decides both the existence and the continuous/discontinuous character of the transition.

What carries the argument

The Gibbs map T that sends a measure to the unique invariant measure of the linearized Fokker–Planck equation; its fixed points are the invariant measures of the nonlinear equation, and the Fréchet derivatives of id−T at those fixed points furnish the linear operators whose spectral conditions decide pitchfork versus saddle-node.

What would settle it

For the one-dimensional asymmetric double-well with quadratic interaction, compute the critical temperature by solving the simultaneous moment-bifurcation and Dawson equations and check whether exactly two new invariant measures appear for temperatures just below that value and none just above, matching the saddle-node prediction.

Watch

Extended reading notes

Core claim

Under standard dissipativity and growth assumptions, continuous phase transitions of the McKean–Vlasov Fokker–Planck equation are pitchfork bifurcations of the Gibbs fixed-point map, while discontinuous transitions are saddle-node bifurcations; when the interaction is quadratic the critical parameter satisfies the generalized Dawson criterion that minus the reciprocal of the critical temperature is a simple eigenvalue of the critical covariance matrix, with bifurcation direction given by the matching eigenvector.

Load-bearing premise

The local theory requires the linearized operator to have a one-dimensional null space; if a critical covariance eigenvalue has multiplicity greater than one, the pitchfork and saddle-node conclusions as stated do not apply.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 5 minor

Summary. The paper studies phase transitions for gradient McKean–Vlasov dynamics in R^n driven by non-convex confining potentials and cooperative interactions, via the nonlinear Fokker–Planck equation and the Gibbs fixed-point map F = id − T. Continuous transitions are characterized by an abstract pitchfork theorem (Theorem 3.1) and a quadratic-interaction specialization (Theorem 3.2 / Corollary 3.5) that recovers a generalized Dawson criterion: −λ⋆⁻¹ is a simple eigenvalue of Cov(ρ⋆), with bifurcation direction given by the corresponding eigenvector. Discontinuous transitions are characterized by a saddle-node theorem (Theorem 3.6). The spectral analysis of the linearized operator K under quadratic interaction is reduced to the covariance matrix via a transformed Hermite basis (Lemma 4.4). Applications include the classical 1-D symmetric double well (continuous/subcritical pitchfork), an asymmetric double well (discontinuous/saddle-node), product four-well models in R² with multiple transitions, a cubic-interaction non-existence result, and a numerical Gaussian-interaction pitchfork.

Significance. The work supplies a systematic bifurcation-theoretic classification of continuous versus discontinuous McKean–Vlasov phase transitions on unbounded space, extending Dawson’s 1-D quadratic criterion to n dimensions and treating discontinuous transitions via saddle-node bifurcation. Strengths include explicit Fréchet derivatives of the Gibbs map (Lemmas 4.1, 4.7), compactness via Fréchet–Kolmogorov–Riesz (Lemma 4.2), a complete finite-rank spectral reduction under (H4), and concrete verification of transversality/pitchfork/non-degeneracy conditions for the double-well and product four-well models. The GHS/Hankel global arguments and the monotonicity inequality in Appendix F are of independent interest. The restriction to simple eigenvalues is stated openly and is standard for Crandall–Rabinowitz; within that scope the central claims are well supported.

minor comments (5)
  1. In the abstract and §1.4 the phrase “despite the dimension-dependent nature of phase transitions” is slightly opaque; a one-sentence clarification that the critical value depends on the spectrum of Cov(ρ⋆) (hence on dimension and well geometry) would help.
  2. Figure 2 caption and the surrounding text refer to “Solutions for Dawson’s criterion (3.8)”; for the asymmetric case the relevant statement is the saddle-node form in Theorem 3.6(III). Aligning the caption with the theorem numbering would avoid confusion.
  3. Example 3.15 (Gaussian interaction) is numerical only; a short remark that the spectral gap of PLKλ⋆ is observed to be ≈0.186 and that the pitchfork coefficients are computed from the same left/right eigenvectors would make the numerical evidence self-contained without requiring the external repository.
  4. A few typographical inconsistencies remain (e.g., “BIFURCA TION THEOR Y” in the title block, occasional missing spaces around “=”, and “sufficiently” with ligature artifacts). A light copy-edit pass would clean these.
  5. The global multiplicity claims for the four-well models (Examples 3.10–3.12) rely on product structure and 1-D GHS; a brief forward pointer that non-product high-dimensional cases are left open (already noted in Remark 3.11) would make the scope of the global statements clearer.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: bifurcation conditions and generalized Dawson criterion are derived from linearization of the fixed-point map, not fitted or self-defined.

full rationale

The paper applies the standard Crandall–Rabinowitz abstract bifurcation theorems (Appendix B) to the fixed-point equation F = id − T of the Gibbs map for the McKean–Vlasov Fokker–Planck equation. The necessary spectral condition (id − λ⋆K noninvertible) is obtained by direct Fréchet differentiation (Lemma 4.1, eqs. 4.1–4.5); under quadratic interaction (H4) the nonzero spectrum of the compact operator K is identified with the eigenvalues of the covariance matrix Cov(ρ⋆) by an explicit matrix representation in the transformed Hermite basis (Lemma 4.4 and Appendix D). Transversality, pitchfork and non-degeneracy conditions reduce to explicit moment formulae that are verified independently for the double-well and product four-well examples via the moment equations, Hankel inequalities and the GHS inequality (Section 5). Numerical critical values for the Gaussian interaction are computed from the same spectral condition, not reverse-engineered. Self-citations supply background context only and are not load-bearing for uniqueness or the central claims. The derivation is therefore self-contained against its own inputs; no step reduces by construction to a fitted parameter or a circular definition.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The work is pure analysis: it imports standard SDE well-posedness and Crandall–Rabinowitz bifurcation, plus domain assumptions (H1)–(H4) on potentials and interactions. No free parameters are fitted to data for the main theorems; numerical critical values in the Gaussian example are outputs of the spectral condition, not inputs. No new physical entities are postulated.

assumptions (5)
  • standard math Crandall–Rabinowitz local bifurcation theorem (and its saddle-node counterpart) in Banach spaces for simple eigenvalues / 1D null spaces (Appendix B, Theorems B.1–B.2).
    Used as the abstract engine for Theorems 3.1, 3.2, 3.6; paper only verifies spectral and nondegeneracy conditions.
  • domain assumption (H1)–(H3): one-sided Lipschitz, polynomial growth, weak hyper-dissipativity of V and W, ensuring well-posedness of the MVSDE and Fokker–Planck equation.
    Stated in §2.2; needed for existence of the Gibbs map T and compactness on weighted L1.
  • domain assumption (H4): attractive symmetric quadratic interaction W(x)=|x|²/2, used for the finite-rank covariance spectral theory and Dawson criterion.
    §3.1; without it the spectrum of K need not reduce to Cov(ρ).
  • domain assumption Existence of a C¹ family of trivial solutions {ρ₀(λ)} through the continuous-transition point (for pitchfork theorems).
    Built into the continuous-transition setup (§2, §3.1); automatic under even V and quadratic W.
  • standard math Ellis–Monroe–Newman GHS inequality (and its strict form when V‴≢0) for even superquadratic potentials with convex derivative on [0,∞).
    Used for global uniqueness/concavity arguments in double-well examples (§5, Appendix E).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Continuity and Discontinuity of McKean-Vlasov Phase Transitions via Bifurcation Theory." pith.science (2026). https://pith.science/paper/B4NAFQ2B

@misc{pith2026260710723,
  author       = {Pith},
  title        = {Pith review of: Continuity and Discontinuity of McKean-Vlasov Phase Transitions via Bifurcation Theory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B4NAFQ2B}},
  note         = {Machine review of arXiv:2607.10723}
}
abstract

It is well known that the McKean-Vlasov stochastic differential equation with a symmetric double-well potential exhibits a continuous phase transition. In contrast, for an asymmetric double-well potential, the system undergoes a discontinuous phase transition, in which an invariant measure abruptly appears in the shallower well and subsequently splits into two as the temperature decreases. In this work, we systematically investigate the types of phase transitions driven by non-convex confining potentials and cooperative interactions in $\mathbb{R}^n$. Our main results consist of a pitchfork bifurcation theorem characterizing continuous phase transitions and a saddle-node bifurcation theorem governing discontinuous phase transitions. In addition, despite the dimension-dependent nature of phase transitions, we are able to extend Dawson's criterion for phase transition points -- originally formulated for quadratic interactions in $1$-dimensional space -- to $n$-dimensional space. This extension directly relates the critical parameter and bifurcation direction to the eigenvalue and eigenvector of the critical covariance matrix, respectively. Finally, we apply our theoretical results to the aforementioned double-well models, to a class of four-well models in two-dimensional Euclidean space that exhibit multiple phase transitions, and to an attractive Gaussian interaction model.

Figures

Figures reproduced from arXiv: 2607.10723 by the authors.

Figure 1
Figure 1. Sketches of symmetric and asymmetric double-well potentials. 1.3. Numerical observations. We illustrate these observed phenomena using the standard proto￾typical examples of one-dimensional symmetric and asymmetric double-well potentials: Vsymm(x) = x 4/4 − x 2/2 and Vasymm(x) = x 4/4 − x 3/3 − x 2/2 (see [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Bifurcation diagrams for mean values of multiple invariant measures obtained by solving the moment bifurcation equation for symmetric and asymmetric double-well potentials: Vsym = x 4/4 − x 2/2, Vasym = x 4/4 − x 3/3 − x 2/2, λ = −2κ/σ2 . The intersection point of solutions of the bifurcation equation (2.3) (black line) and Dawson’s criterion (3.8) (red dashed line) is exactly the (continuous or discontinuous) phase… view at source ↗
Figure 3
Figure 3. Multiple phase transitions for V (x1, x2) = x 4 1 − 2x 2 1 + x 4 2 /4 − x 2 2 /2. Red dashed arrows denote the first phase transition at temperature σ (1) ⋆ ; brown arrows denote the second phase transitions at temperature σ (2) ⋆ . Remark 3.11. To ensure that the null space of the linear part of the bifurcation equation is 1-dimensional, V1 6= V2 is assumed. It is numerically observed that if V1 = V2, all phase tra… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Graphs of hxiρ and id with respect to m as σ varies, with κ = 1. In the symmetric double-well case, the graph is centrally symmetric with respect to the point (0, 0). When 0 < σ < σ⋆, hxiρ has three fixed points. When σ = σ⋆, the graph of hxiρ is tangent to id at the o…
Figure 5
Figure 5. Figure 5: Graphs of hxiρ and id with respect to m as σ varies, with κ = 1. In the asymmetric double-well case, the graph is centrally symmetric with respect to the point (−2/27, 1/3). When 0 < σ < σ⋆, hxiρ has three fixed points; when σ = σ⋆, the graph of hxiρ is tangent to id; …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 3 linked inside Pith

  1. [1]

    Andersson and P

    A. Andersson and P. Sjögren. Ornstein-Uhlenbeck theory in finite dimension, 2012. Online. https://www.math. chalmers.se/~donnerda/OU.pdf

    2012

  2. [2]

    Balasubramanian, S

    K. Balasubramanian, S. Banerjee, and P. Rigollet. On the structure of stationary solutions to McKean-Vlasov equations with applications to noisy transformers, 2025. arXiv:2510.20094

    arXiv 2025

  3. [3]

    Bao and J

    J. Bao and J. Wang. Stationary distributions of McKean-Vlasov SDEs with jumps: Existence, multiplicity and uniqueness. Stochastic Process. Appl. 198, Article 104967, 2026

    2026

  4. [4]

    K. Bashiri. On the long-time behaviour of McKean-Vlasov paths. Electron. Commun. Probab. 25, Paper No. 52, 14 pp., 2020

    2020

  5. [5]

    Buffoni and J

    B. Buffoni and J. Toland. Analytic Theory of Global Bifurcation . Princeton Series in Applied Mathematics. Princeton University Press, Princeton, NJ, 2003

    2003

  6. [6]

    J. A. Carrillo, R. S. Gvalani, G. A. Pavliotis, and A. Schlichting. Long-time behaviour and phase transitions for the McKean-Vlasov equation on the torus. Arch. Ration. Mech. Anal. 235(1):635–690, 2020

    2020

  7. [7]

    J. A. Carrillo, R. J. McCann, and C. Villani. Kinetic equilibration rates for granular media and related equations: entropy dissipation and mass transportation estimates. Rev. Mat. Iberoam. 19(3):971–1018, 2003

    2003

  8. [8]

    Chayes and V

    L. Chayes and V. Panferov. The McKean-Vlasov equation in finite volume. J. Stat. Phys. 138(1-3):351–380, 2010

    2010

Show all 47 references
  1. [9]

    Q. Cormier. On the stability of the invariant probability measures of McKean-Vlasov equations. Ann. Inst. Henri Poincaré Probab. Stat. 61(4):2405–2429, 2025

    2025

  2. [10]

    M. G. Crandall and P. H. Rabinowitz. Bifurcation from simple eigenvalues. J. Funct. Anal. 8:321–340, 1971

    1971

  3. [11]

    M. G. Crandall and P. H. Rabinowitz. Bifurcation, perturbation of simple eigenvalues and linearized stability. Arch. Ration. Mech. Anal. 52:161–180, 1973

    1973

  4. [12]

    Cormier, E

    Q. Cormier, E. Tanré, and R. Veltz. Hopf bifurcation in a mean-field model of spiking neurons. Electron. J. Probab. 26, Paper No. 121, 40 pp., 2021

    2021

  5. [13]

    D. A. Dawson. Critical dynamics and fluctuations for a mean-field model of cooperative behavior. J. Stat. Phys. 31(1):29–85, 1983

    1983

  6. [14]

    D. A. Dawson and J. Gärtner. Large deviations, free energy functional and quasi-potential for a mean field model of interacting diffusions. Mem. Amer. Math. Soc. 78(398):iv+94, 1989

    1989

  7. [15]

    Deimling

    K. Deimling. Nonlinear Functional Analysis . Springer-Verlag, Berlin, 1985

    1985

  8. [16]

    M. G. Delgadino, R. S. Gvalani, and G. A. Pavliotis. On the diffusive-mean field limit for weakly interacting diffusions exhibiting phase transitions. Arch. Ration. Mech. Anal. 241(1):91–148, 2021

    2021

  9. [17]

    M. G. Delgadino, R. S. Gvalani, G. A. Pavliotis, and S. A. Smith. Phase transitions, logarithmic Sobolev inequalities, and uniform-in-time propagation of chaos for weakly interacting diffusions. Comm. Math. Phys. 401(1):275–323, 2023. MCKEAN-VLASOV PHASE TRANSITIONS VIA BIFURC...

    2023

  10. [18]

    R. S. Ellis. Concavity of magnetization for a class of even ferromagnets. Bull. Amer. Math. Soc. 81(5):925–929, 1975

    1975

  11. [19]

    R. S. Ellis, J. L. Monroe, and C. M. Newman. The GHS and other correlation inequalities for a class of even ferromagnets. Comm. Math. Phys. 46(2):167–182, 1976

    1976

  12. [20]

    Gerber, R

    N. Gerber, R. Gvalani, M. Hairer, G. Pavliotis, and A. Schlichting. Formation of clusters and coarsening in weakly interacting diffusions, 2025. arXiv:2510.17629

    2025 arXiv

  13. [21]

    S. N. Gomes, S. Kalliadasis, G. A. Pavliotis, and P. Yatsyshin. Dynamics of the Desai-Zwanzig model in multiwell and random energy landscapes. Phys. Rev. E. 99(3):032109, 2019

    2019

  14. [22]

    Hanche-Olsen and H

    H. Hanche-Olsen and H. Holden. The Kolmogorov-Riesz compactness theorem. Expo. Math. 28(4):385–394, 2010

    2010

  15. [23]

    Herrmann and J

    S. Herrmann and J. Tugaut. Non-uniqueness of stationary measures for self-stabilizing processes. Stochastic Process. Appl. 120(7):1215–1246, 2010

    2010

  16. [24]

    W. Hong, S. Hu and W. Liu. McKean-Vlasov SDE and SPDE with locally monotone coefficients. Ann. Appl. Probab. 34(2):2136–2189, 2024

    2024

  17. [25]

    Jordan, D

    R. Jordan, D. Kinderlehrer, and F. Otto. The variational formulation of the Fokker-Planck equation. SIAM J. Math. Anal. 29(1):1–17, 1998

    1998

  18. [26]

    R. B. Kellogg, T. Y. Li, and J. Yorke. A constructive proof of the Brouwer fixed-point theorem and computational results. SIAM J. Numer. Anal. 13(4):473–483, 1976

    1976

  19. [27]

    Kielhöfer

    H. Kielhöfer. Bifurcation Theory, volume 156 of Applied Mathematical Sciences . Springer, New York, second edition, 2012. An introduction with applications to partial differential equations

    2012

  20. [28]

    Leimkuhler, R

    B. Leimkuhler, R. Lohmann, G. A. Pavliotis, and P. A. Whalley. Cluster formation in diffusive systems, 2025. arXiv:2510.25034

    2025 arXiv

  21. [29]

    C. Liu, B. Qu, J. Yao, and Y. Zhi. Connecting orbits in cooperative McKean-Vlasov SDEs, 2026. arXiv:2407.18853v2

    2026

  22. [30]

    W. Liu, L. Wu, and C. Zhang. Long-time behaviors of mean-field interacting particle systems related to McKean- Vlasov equations. Comm. Math. Phys. 387(1):179–214, 2021

    2021

  23. [31]

    Lu ￿ con and C

    E. Lu ￿ con and C. Poquet. Emergence of oscillatory behaviors for excitable systems with noise and mean-field interaction: a slow-fast dynamics approach. Comm. Math. Phys. 373(3):907–969, 2020

    2020

  24. [32]

    Lu ￿ con and C

    E. Lu ￿ con and C. Poquet. Periodicity and longtime diffusion for mean field systems in Rd. Ann. Inst. Henri Poincaré Probab. Stat. 61(4):2430–2467, 2025

    2025

  25. [33]

    H. P. McKean, Jr. Propagation of chaos for a class of non-linear parabolic equations. In Stochastic Differential Equations (Lecture Series in Differential Equations, Session 7, Catholic Univ., 1967) , volume 7 of Lecture Series in Differential Equations , pages 41–57. Air Forc...

    1967

  26. [34]

    Mischler, C

    S. Mischler, C. Quiñinao, and J. D. Touboul. On a kinetic Fitzhugh-Nagumo model of neuronal network. Comm. Math. Phys. 342(3):1001–1042, 2016

    2016

  27. [35]

    B. M. Mladek, D. Gottwald, G. Kahl, M. Neumann, and C. N. Likos. Formation of polymorphic cluster phases for a class of models of purely repulsive soft spheres. Phys. Rev. Lett. 96(4):045701, 2006

    2006

  28. [36]

    Monmarché

    P. Monmarché. Long-time propagation of chaos and exit times for metastable mean-field particle systems, 2025. arXiv:2503.00157

    2025 arXiv

  29. [37]

    J. E. Osborn. Spectral approximation for compact operators. Math. Comput. 29(131):712–725, 1975

    1975

  30. [38]

    B. Qu, J. Yao, and Y. Zhi. Global dynamics of granular media equations via stochastic order, 2026. arXiv:2411.01364v2

    2026

  31. [39]

    Scheutzow

    M. Scheutzow. Periodic behavior of the stochastic Brusselator in the mean-field limit. Probab. Theory Relat. Fields 72(3):425–462, 1986

    1986

  32. [40]

    Shalova and A

    A. Shalova and A. Schlichting. Solutions of stationary McKean-Vlasov equation on a high-dimensional sphere and other Riemannian manifolds. Adv. Nonlinear Anal. 15(1):20250141, 48 pp., 2026

    2026

  33. [41]

    J. A. Shohat and J. D. Tamarkin. The Problem of Moments , Vol. I of American Mathematical Society Mathe- matical Surveys . American Mathematical Society, New York, 1943

    1943

  34. [42]

    F. H. Stillinger. Phase transitions in the Gaussian core system. J. Chem. Phys. 65(10):3968–3974, 1976

    1976

  35. [43]

    Sznitman

    A.-S. Sznitman. Topics in propagation of chaos. In École d’Été de Probabilités de Saint-Flour XIX—1989 , volume 1464 of Lecture Notes in Math. pages 165–251. Springer, Berlin, 1991

    1989

  36. [44]

    Y. Tamura. On asymptotic behaviors of the solution of a nonlinear diffusion equation. J. Fac. Sci. Univ. Tokyo Sect. IA Math. 31(1):195–221, 1984

    1984

  37. [45]

    J. Tugaut. Convergence to the equilibria for self-stabilizing processes in double-well landscape. Ann. Probab. 41(3A):1427–1460, 2013

    2013

  38. [46]

    S.-Q. Zhang. A local bifurcation theorem for McKean-Vlasov diffusions. J. Funct. Anal. 289(11), Article 111144, 60 pp., 2025

    2025

  39. [47]

    S.-Q. Zhang. Local convergence near equilibria for distribution dependent SDEs, 2025. arXiv:2501.04313. 50 JUNLANG HU AND ZHENXIN LIU J. Hu: School of Mathematical Sciences, Dalian University of Technology, Dalian 116024, P. R. China Email address : junlanghu@hotmail.com Z. Li...

    2025

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.