Theoretical determination of the Kardar–Parisi–Zhang universality class above one dimension: critical exponents, with the (2+1)-dimensional class as the unsolved core

The Kardar–Parisi–Zhang (KPZ) universality class governs stochastic interface growth and is exactly solved only in (1+1) dimensions. A 2015 high-precision simulation of the restricted-solid-on-solid model on two-dimensional substrates (Pagnani and Parisi, Phys. Rev. E 92, 010101(R)) fixed the roughness exponent at $\chi = 0.3869(4)$ and the steady-state finite-size correction exponent at $\omega = 0.56(5)$, excluded the Kim–Kosterlitz conjecture $\chi = 2/5$ at over $32\sigma$, and characterized the height statistics only through three cumulants — while stating that the quantification of KPZ critical exponents at any substrate dimension $d>1$ remains an open challenge. The surviving open problem attached to that paper is an exact or controlled theoretical determination of the KPZ critical exponents in $d>1$, with the (2+1)-dimensional class as its unsolved core, including the reconciliation of nonperturbative renormalization-group estimates ($\chi \approx 0.33$) with high-precision numerics. Two related items are separately recognized as open in the surrounding literature: a theory of the steady-state correction exponent $\omega$ and of the observed second-order structure $1 + B L^{-\omega} + C L^{-2\omega}$, and an analytical form of the full universal steady-state height-fluctuation distribution in (2+1) dimensions. No resolution of these questions is found in the literature through 2026: numerical characterizations have matured, several unproven conjectures compete, and explicit statements of openness appear in works from 2015 to 2026.

Background

The Kardar–Parisi–Zhang (KPZ) equation (Kardar, Parisi, and Zhang, Phys. Rev. Lett. 56, 889 (1986)) is the minimal continuum model of a stochastically growing interface, $\partial_t h(\mathbf{r},t) = \nu \nabla^2 h + \tfrac{\lambda}{2}|\nabla h|^2 + \eta(\mathbf{r},t)$, where $h$ is the interface height, $\nu$ the diffusion coefficient, $\lambda$ the strength of the growth nonlinearity, and $\eta$ Gaussian white noise. Its scale-invariant solutions define the KPZ universality class, which collects lattice growth models, directed polymers in random media (DPRM — the polymer's free-energy fluctuations map onto KPZ height fluctuations), and, more recently, driven-dissipative condensates. In $d$ substrate dimensions the class is characterized by the roughness exponent $\chi$ and dynamic exponent $z$, which fix how interface fluctuations scale with system size and time ($\beta = \chi/z$ is the growth exponent; the numerical literature often denotes the roughness exponent by $\alpha$, and $\alpha \equiv \chi$ throughout this statement); Galilean (tilt) invariance implies the scaling relation $\chi + z = 2$. In $1+1$ dimensions the class is exactly solved — $\chi = 1/2$, $z = 3/2$, with height fluctuations following Tracy–Widom distributions from random matrix theory — but in every higher dimension it is a strongly coupled, interacting fixed point with no exact or controlled solution.

The source paper (A. Pagnani and G. Parisi, Phys. Rev. E 92, 010101(R) (2015); arXiv:1611.08445) addressed this in $d = 2$ numerically. It simulated the restricted-solid-on-solid (RSOS) growth model — a lattice KPZ-class realization in which the height difference between neighboring sites is restricted to at most one lattice unit — on square substrates of linear size $L$ up to 480, using multispin coding (a bit-parallel simulation technique that packs many independent samples into one machine word, giving extremely high statistics). Measuring the steady-state height cumulants $w_2, w_3, w_4$ (the second, third, and fourth centered moments of the height) and fitting them simultaneously with the finite-size ansatz $w_k \propto L^{k\chi}(1 + B L^{-\omega} + C L^{-2\omega})$, it obtained $\chi = 0.3869(4)$ and $\omega = 0.56(5)$. This excluded the long-standing Kim–Kosterlitz conjecture $\chi_{d=2} = 2/5$ (J. M. Kim and J. M. Kosterlitz, Phys. Rev. Lett. 62, 2289 (1989)) at more than $32$ standard deviations and showed the steady-state height statistics to be strongly non-Gaussian, quantified only through the cumulant ratios $R_3 = w_3/w_2^{3/2}$ and $R_4 = w_4/w_2^2$. The paper reports that the nonperturbative renormalization-group (NPRG — a functional renormalization-group scheme that evolves an effective action over momentum scales and can reach strong-coupling fixed points at the cost of uncontrolled truncation errors) estimates $\chi = 0.33$, $\omega = 0.7$ are "still far from our numerical estimate", and states that at any $d > 1$ the quantification of the critical exponents is still an open challenge.

The surrounding literature frames the same gaps. The Halpin–Healy–Takeuchi review (J. Stat. Phys. 160, 794 (2015)) describes the (2+1)-dimensional limit distributions as numerically well characterized — geometry-dependent universal classes with stationary variance $\langle\xi^2\rangle = 0.464$ and skewness $s_0 = 0.244$ — while explicitly hoping that field theorists will someday craft a renormalization-group portrait of the entire limit distribution. Later works (e.g., T. J. Oliveira, Phys. Rev. E 106, L062103 (2022); Gomes-Filho et al., Entropy 26, 260 (2024)) still describe the exact exponents in any $d$ and the height distributions above one dimension as central open questions. What is missing is not numerics but theory: exact values, controlled derivations, or first-principles explanations of the exponents, the correction structure, and the full distribution in the (2+1)-dimensional class.

Problem Statement

Provide an exact or controlled first-principles theoretical determination of the critical exponents of the Kardar–Parisi–Zhang universality class above one substrate dimension — the open challenge stated by the source paper, which reports that at any $d>1$ the quantification of the critical exponents is still an open challenge — with the (2+1)-dimensional class as its unsolved core. The scientific objective is to replace the current purely numerical characterization of the exponents by a first-principles one:

1. Critical exponents. Determine the KPZ roughness and dynamic exponents ($\chi$, $z$) of the (2+1)-dimensional class — exactly, or by a controlled nonperturbative method with quantified errors — and reconcile the analytical approximations, in particular the NPRG estimates $\chi \approx 0.33$ and $\omega \approx 0.7$ in $d = 2$, with the high-precision numerical values ($\chi = 0.3869(4)$ and the competing estimates $\alpha = 0.390(4)$ and $\alpha = 0.393(4)$ from the DLC model of Kelling and Ódor, Phys. Rev. E 84, 061150 (2011)). This includes assessing the proposed closed forms — the conjecture $\beta_d = 7/(8d+13)$ (giving $\chi(d=2) = 7/18$ via $\beta = \chi/z$ and $\chi + z = 2$) and the golden-ratio fractal-dimension ansatz $\alpha = 0.381966\ldots$ — against a controlled derivation. The Kim–Kosterlitz value $\chi = 2/5$, numerically excluded at more than $32\sigma$ by the source paper, stands as historical context rather than a live candidate.

Theoretically determine, from first principles, the critical exponents of the (2+1)-dimensional KPZ class; a determination valid for general $d>1$ answers the source's open challenge in its full generality. The following related open items are separately recognized in the literature:

2. Correction exponent. The microscopic origin and the universality (or model-dependence) of the steady-state finite-size correction exponent $\omega$: whether $\omega$ is a universal KPZ quantity, whether it is related to $\chi$ and $z$, and whether the observed second-order structure $1 + B L^{-\omega} + C L^{-2\omega}$ in the height cumulants reflects a physical feature of the class or a model-dependent fitting scheme — the strength and structure of the corrections are known to depend on microscopic details, and a competing "generalized KPZ ansatz" attributes finite-time correction exponents to multiples of $\beta$ rather than to the steady-state $\omega = 0.56(5)$.

3. Height distribution. The full universal steady-state height-fluctuation probability distribution in $d = 2$ — beyond the first three cumulants and their ratios currently measured — whose analytical derivation, and whose geometry-dependent universality (the stationary, flat, radial, and cylindrical classes observed across lattice models of the KPZ class), remain open.

References

1. A. Pagnani and G. Parisi, "Numerical estimate of the Kardar-Parisi-Zhang universality class in (2+1) dimensions", Phys. Rev. E 92, 010101(R) (2015). DOI: 10.1103/PhysRevE.92.010101; arXiv:1611.08445. https://arxiv.org/abs/1611.08445
2. Mehran Kardar, Giorgio Parisi, and Yi-Cheng Zhang, "Dynamic Scaling of Growing Interfaces", Phys. Rev. Lett. 56, 889 (1986). DOI: 10.1103/PhysRevLett.56.889. https://doi.org/10.1103/PhysRevLett.56.889
3. Jin Min Kim and J. M. Kosterlitz, "Growth in a restricted solid-on-solid model", Phys. Rev. Lett. 62, 2289 (1989). DOI: 10.1103/PhysRevLett.62.2289. https://doi.org/10.1103/PhysRevLett.62.2289
4. Timothy Halpin-Healy and Kazumasa A. Takeuchi, "A KPZ Cocktail-Shaken, not Stirred: Toasting 30 Years of Kinetically Roughened Surfaces", J. Stat. Phys. 160, 794 (2015). DOI: 10.1007/s10955-015-1282-1; arXiv:1505.01910. https://doi.org/10.1007/s10955-015-1282-1
5. Léonie Canet, Hugues Chaté, Bertrand Delamotte, and Nicolás Wschebor, "Nonperturbative renormalization group for the Kardar-Parisi-Zhang equation: General framework and first applications", Phys. Rev. E 84, 061128 (2011). DOI: 10.1103/PhysRevE.84.061128; arXiv:1107.2289. https://arxiv.org/abs/1107.2289
6. Thomas Kloss, Léonie Canet, and Nicolás Wschebor, "Nonperturbative renormalization group for the stationary Kardar-Parisi-Zhang equation: Scaling functions and amplitude ratios in 1+1, 2+1, and 3+1 dimensions", Phys. Rev. E 86, 051124 (2012). DOI: 10.1103/PhysRevE.86.051124. https://doi.org/10.1103/PhysRevE.86.051124
7. T. J. Oliveira, S. G. Alves, and S. C. Ferreira, "Kardar-Parisi-Zhang universality class in (2+1) dimensions: Universal geometry-dependent distributions and finite-time corrections", Phys. Rev. E 87, 040102(R) (2013). DOI: 10.1103/PhysRevE.87.040102. https://doi.org/10.1103/PhysRevE.87.040102
8. Sidiney G. Alves, Tiago J. Oliveira, and Silvio C. Ferreira, "Origins of scaling corrections in ballistic growth models", Phys. Rev. E 90, 052405 (2014). DOI: 10.1103/PhysRevE.90.052405; arXiv:1409.2460. https://arxiv.org/abs/1409.2460
9. Sidiney G. Alves and Silvio C. Ferreira, "Scaling, cumulant ratios, and height distribution of ballistic deposition in 3+1 and 4+1 dimensions", Phys. Rev. E 93, 052131 (2016). DOI: 10.1103/PhysRevE.93.052131. https://doi.org/10.1103/PhysRevE.93.052131
10. Jeffrey Kelling, Géza Ódor, and Sibylle Gemming, "Universality of (2+1)-dimensional restricted solid-on-solid models", Phys. Rev. E 94, 022107 (2016). DOI: 10.1103/PhysRevE.94.022107; arXiv:1605.02620. https://arxiv.org/abs/1605.02620
11. Tapas Singha and Malay K. Nandy, "Steady-state skewness and kurtosis from renormalized cumulants in (2+1)-dimensional stochastic surface growth", J. Stat. Mech. (2016) 103204. DOI: 10.1088/1742-5468/2016/10/103204; arXiv:1609.02547. https://arxiv.org/abs/1609.02547
12. Márcio S. Gomes-Filho, André L. A. Penna, and Fernando A. Oliveira, "The Kardar-Parisi-Zhang exponents for the 2+1 dimensions", Results in Physics 26, 104435 (2021). DOI: 10.1016/j.rinp.2021.104435; arXiv:2006.11417. https://arxiv.org/abs/2006.11417
13. Tiago J. Oliveira, "Kardar-Parisi-Zhang universality class in (d+1)-dimensions", Phys. Rev. E 106, L062103 (2022). DOI: 10.1103/PhysRevE.106.L062103; arXiv:2212.03847. https://arxiv.org/abs/2212.03847
14. Ismael S. S. Carrasco and Tiago J. Oliveira, "One-point height fluctuations and two-point correlators of (2+1) cylindrical KPZ systems", Phys. Rev. E 107, 064140 (2023). DOI: 10.1103/PhysRevE.107.064140; arXiv:2305.01062. https://arxiv.org/abs/2305.01062
15. Ismael S. S. Carrasco and Tiago J. Oliveira, "Dimensional crossover in Kardar-Parisi-Zhang growth", Phys. Rev. E 109, L042102 (2024). DOI: 10.1103/PhysRevE.109.L042102. https://doi.org/10.1103/PhysRevE.109.L042102
16. Márcio S. Gomes-Filho, Pablo de Castro, Danilo B. Liarte, and Fernando A. Oliveira, "Restoring the Fluctuation–Dissipation Theorem in Kardar–Parisi-Zhang Universality Class through a New Emergent Fractal Dimension", Entropy 26(3), 260 (2024). DOI: 10.3390/e26030260; arXiv:2402.15576. https://arxiv.org/abs/2402.15576
17. J. M. Marcos, J. J. Meléndez, R. Cuerno, and J. J. Ruiz-Lorenzo, "On the upper critical dimension of the KPZ universality class: KPZ and related equations on a fully connected graph", arXiv:2603.02000 (2026). https://arxiv.org/abs/2603.02000
18. Simon Widmann, Siddhartha Dam, Johannes Düreth, Christian G. Mayer, Romain Daviet, Carl Philipp Zelle, David Laibacher, Monika Emmerling, Martin Kamp, Sebastian Diehl, Simon Betzold, Sebastian Klembt, and Sven Höfling, "Observation of Kardar-Parisi-Zhang universal scaling in two dimensions", Science 392, 221 (2026). DOI: 10.1126/science.aeb4154; arXiv:2506.15521. https://doi.org/10.1126/science.aeb4154
19. Jeffrey Kelling and Géza Ódor, "Extremely rough phase in surface growth", Phys. Rev. E 84, 061150 (2011). DOI: 10.1103/PhysRevE.84.061150. https://doi.org/10.1103/PhysRevE.84.061150