Talenti-type Inequalities in PDE Analysis

• Talenti-type inequalities are fundamental functional inequalities that compare solutions of elliptic and parabolic PDEs through geometric symmetrization, ensuring optimal analytic bounds.
• They provide quantitative stability and rigidity results, meaning that near-equality forces the domain and data to be close to symmetric models like balls.
• Extensions to weighted, Riemannian, and fractional settings enable applications in sharp spectral gap estimates, Sobolev inequalities, and probabilistic interpretations.

Talenti-type inequalities are a central class of sharp functional inequalities and symmetrization comparison results in the theory of elliptic and parabolic partial differential equations, geometric analysis, and spectral theory. These inequalities quantify the effect of geometric symmetrization (notably, Schwarz or spherical rearrangement) on the solutions to PDEs and on associated energies and eigenvalues, providing both analytic bounds and geometric rigidity/stability. Talenti-type theorems originally appeared for solutions to elliptic Dirichlet and Robin problems in Euclidean domains, but robust generalizations now exist across weighted, Riemannian, metric-measure, and even singular geometric frameworks.

1. Core Principles: Rearrangement and Symmetrization

At the heart of Talenti-type inequalities is the observation that rearranging a source function $f$ by its Schwarz (radially decreasing) symmetrization allows one to compare solutions to elliptic PDEs on arbitrary domains with those on balls, where optimal comparison is achieved.

Schwarz Rearrangement and Symmetrization

• For a measurable $h \geq 0$ on $\Omega \subset \mathbb{R}^n$, the distribution function is $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$, and the decreasing rearrangement is $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$, $0 < s < |\Omega|$.
• The Schwarz rearrangement $h^\sharp(x) = h^*(\omega_n |x|^n)$, with $\omega_n$ the volume of the unit ball, is radially symmetric, non-increasing, and equimeasurable with $h$.
• This framework ensures equidistribution of superlevel measures and norm preservation for rearrangement-invariant norms: $$\|h\|_{L^p(\Omega)} = \|h^\sharp\|_{L^p(\Omega^\sharp)}$$.

The Classical Dirichlet Problem and Talenti's Theorem

Given a bounded domain $h \geq 0$0 and $h \geq 0$1, let $h \geq 0$2 solve $h \geq 0$3 in $h \geq 0$4, $h \geq 0$5 on $h \geq 0$6. Let $h \geq 0$7 solve $h \geq 0$8 in the ball $h \geq 0$9 with $\Omega \subset \mathbb{R}^n$0 on $\Omega \subset \mathbb{R}^n$1. Then

$\Omega \subset \mathbb{R}^n$2

with equality only for balls and radially symmetric data. Corresponding inequalities hold for rearrangement-invariant function spaces, and for gradient norms $\Omega \subset \mathbb{R}^n$3, $\Omega \subset \mathbb{R}^n$4 (Amato et al., 2023).

These results generalize with weighted measures, more general operators (e.g., $\Omega \subset \mathbb{R}^n$5-Laplacian), and alternative geometric settings.

2. Quantitative and Rigidity Enhancements

Beyond the classical comparison, modern research establishes quantitative stability and rigidity results: near-equality in Talenti-type inequalities forces the underlying domain and data close (in optimal geometric/topological sense) to symmetric models (balls, cones, or suspensions), and yields explicit asymmetry bounds.

Quantitative Stability

Let $\Omega \subset \mathbb{R}^n$6 denote the Fraenkel asymmetry of $\Omega \subset \mathbb{R}^n$7, measuring the normalized deviation (in Lebesgue measure) from a ball. There exist explicit constants and exponents such that

$\Omega \subset \mathbb{R}^n$8

for solutions $\Omega \subset \mathbb{R}^n$9, $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$0 as above (Amato et al., 2023). This yields effective bounds: optimality of the Talenti comparison identifies domains and data near a ball and its rearrangement.

Rigidity

Equality in Talenti-type inequalities throughout an interval (or at a point for monotone solutions) forces geometric rigidity: the domain is (up to null sets) a ball, and the data are radial (Amato et al., 2023, Amato et al., 14 Nov 2025). For many settings (notably RCD spaces, see below), this rigidity fully characterizes spaces with equality as cones or spherical suspensions (Wu, 2024, Wu, 8 Jun 2025).

3. Variational, Spectral, and Geometric Extensions

Talenti-type principles underlie a broad class of functional inequalities and spectral estimates, applicable across manifold and measure settings.

Weighted and Generalized Symmetrization

Talenti symmetrization extends to arbitrary weighted measures $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$1 (absolutely continuous with respect to Lebesgue). For sufficiently regular weights and suitable isoperimetric minimizers among level sets (e.g., balls, half-spaces), one establishes the weighted Pólya–Szegő inequality

$\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$2

and thus spectral inequalities for, e.g., the first $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$3-eigenvalue of the weighted Dirichlet $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$4-Laplacian: $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$5 These comparisons encapsulate classical Euclidean, Gaussian, log-concave, homogeneous, and anisotropic settings (Bartoli et al., 28 May 2026).

Higher-Order, Parabolic, and Fractional Counterparts

• Higher-order Sobolev/Rellich inequalities: Iteration of Talenti's principle yields consistent sharp constants for higher-order operators (bi-Laplacian and beyond) on manifolds supporting sharp isoperimetry (Farkas et al., 23 Sep 2025).
• Parabolic Talenti: For the heat equation, rearranging in space yields order/comparison at fixed time, but full time-space maximality cannot be achieved under global integral constraints (Idriss, 2022).
• Fractional Laplacian: The pointwise Talenti comparison fails universally in the fractional (nonlocal) radial setting, with only mass concentration inequalities or alternative boundary Talenti inequalities surviving in specific dimensions and regimes (Karrouchi et al., 2024). The critical order $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$6 marks a transition in comparison direction.

Extensions to Non-Euclidean Spaces

Talenti-type inequalities have been proved in settings of nonnegative Ricci curvature, Cartan-Hadamard manifolds, Alexandrov surfaces, and—most notably—metric measure spaces with synthetic Ricci bounds, such as $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$7 spaces.

• RCD($\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$8) spaces (Ricci lower bound, dimension upper bound): The Dirichlet solution $\mu_h(t) = |\{ x \in \Omega : h(x) > t \}|$9 to $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$0 on $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$1, compared with the corresponding solution in the 1D model space $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$2, satisfies $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$3 pointwise, with $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$4 solving the symmetrized model problem. Rigidity and stability precisely characterize the model spaces (spherical suspensions) for which equality is achieved (Mondino et al., 2020, Wu, 2024, Wu, 8 Jun 2025).

4. Boundary Conditions: Robin, Neumann, and Multiply Connected Domains

Robin Problems and Lorentz Space Comparison

For Robin boundary data, Talenti-type inequalities become dimension-sensitive. In two dimensions and for constant source, the pointwise rearrangement inequality persists: $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$5 However, in higher dimensions only Lorentz-norm comparisons can be proved: $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$6 for all admissible $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$7, with precise threshold exponents characterizing the range (Alvino et al., 2019). Both sharpness and open problems exist: pointwise comparisons and $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$8-norms remain open for $h^*(s) = \inf\{ t \geq 0 : \mu_h(t) < s \}$9.

Quantitative versions, with explicit asymmetry control, are available: $0 < s < |\Omega|$0 with matching rigidity: equality implies radial symmetry (Amato et al., 14 Nov 2025).

Neumann Problems

Analogous symmetrization theorems hold for Neumann problems, but an additional boundary matching condition (on $0 < s < |\Omega|$1 or $0 < s < |\Omega|$2) is required to ensure meaningful comparison (Celentano et al., 2024). In two dimensions and constant source, this recovers the classical pointwise result; higher dimensions again yield only Lorentz-norm comparisons, sensitive to the function class.

Multiply Connected Domains

Talenti-type inequalities are established for Robin-type problems with inner Dirichlet and outer Robin data, even on multiply connected domains. Here, the symmetric model is an annulus or union of balls ("radial annulus with holes"). The main comparison theorems remain of Lorentz norm type, sharpening to pointwise rearrangement only for $0 < s < |\Omega|$3 (Barbato et al., 8 Apr 2025).

5. Applications: Sharp Constants, Spectral Gaps, and Probabilistic Interpretations

Sobolev, Faber–Krahn, and Saint-Venant

• Talenti’s principle yields the sharp Sobolev inequality in $0 < s < |\Omega|$4 with optimal constant: $0 < s < |\Omega|$5 with equality for Aubin–Talenti bubbles (Dolbeault et al., 2013, Kniefacz et al., 2019, Farkas et al., 23 Sep 2025).
• The Faber–Krahn inequality for the first Dirichlet eigenvalue follows directly, as does the torsional rigidity (Saint–Venant) inequality and various reverse Hölder inequalities for eigenfunctions (Chen et al., 2021, Mondino et al., 2020, Wu, 8 Jun 2025).

Geometric and Probabilistic Interpretations

• On $0 < s < |\Omega|$6 or Riemannian geometric spaces, the corresponding comparison solution is interpretable as an expected Brownian exit time. The Talenti–type comparison associates sharper estimates for the mean exit time from arbitrary domains relative to the symmetric model domain (Mondino et al., 2020).
• In the hyperbolic or more general non-Euclidean context, the Talenti principle carries forward after a compensating subtraction of a (non-attainable) Hardy/Poincaré term, maintaining the same optimal constants (Nguyen, 2018).

6. Limitations, Counterexamples, and Future Directions

• Boundary and Fractional Laplacians: For fractional Laplacians, the classical Talenti comparison fails even in the radial case for $0 < s < |\Omega|$7. Instead, only mass-concentration or certain boundary (fractional normal derivative) inequalities survive under additional hypotheses (Karrouchi et al., 2024).
• Parabolic Maximality: Full time-space rearrangement maximality does not exist under global constraints, even though Talenti–type inequalities hold at each fixed time (Idriss, 2022).
• Open Problems: Extension of pointwise or $0 < s < |\Omega|$8-Talenti inequalities in the Robin setting for $0 < s < |\Omega|$9 remains unresolved (Alvino et al., 2019). Optimal exponent thresholds in Lorentz spaces, and sharpened stability exponents, are topics of active investigation (Amato et al., 2023, Acampora et al., 10 Mar 2025).

7. Schematic Table of Principal Inequality Types

Setting	Main Comparison Type	Explicit Constant/Sharpness
Dirichlet, Euclidean	pointwise, norm (all $h^\sharp(x) = h^*(\omega_n |x|^n)$0)	Yes, Aubin–Talenti
Robin, $h^\sharp(x) = h^*(\omega_n |x|^n)$1, $h^\sharp(x) = h^*(\omega_n |x|^n)$2	pointwise	Yes
Robin, $h^\sharp(x) = h^*(\omega_n |x|^n)$3	Lorentz norm ($h^\sharp(x) = h^*(\omega_n |x|^n)$4)	Thresholds only, sharp
Weighted ($h^\sharp(x) = h^*(\omega_n |x|^n)$5-Laplacian)	norm, spectrum	Yes
RCD($h^\sharp(x) = h^*(\omega_n |x|^n)$6) metric spaces	pointwise/gradient/norm	Model space, sharp/rigid
Fractional Laplacian ($h^\sharp(x) = h^*(\omega_n |x|^n)$7)	mass-concentration, not pointwise	No
Neumann/Multiply Connected	Lorentz norm with boundary match	Yes (dimension/form-dependent)

• "A Talenti comparison result for solutions to elliptic problems with Robin boundary conditions" (Alvino et al., 2019)
• "The Talenti comparison result in a quantitative form" (Amato et al., 2023)
• "A quantitative Talenti-type comparison result with Robin boundary conditions" (Amato et al., 14 Nov 2025)
• "Spectral inequalities for weighted $h^\sharp(x) = h^*(\omega_n |x|^n)$8-Laplacians via Talenti symmetrization" (Bartoli et al., 28 May 2026)
• "A Talenti-type comparison theorem for $h^\sharp(x) = h^*(\omega_n |x|^n)$9 spaces and applications" (Mondino et al., 2020)
• "A Talenti-type comparison theorem for the $\omega_n$0-Laplacian on $\omega_n$1 spaces and some applications" (Wu, 2024)
• "Almost rigidity of the Talenti-type comparison theorem on $\omega_n$2 space" (Wu, 8 Jun 2025)
• "A Talenti comparison result for a class of Neumann boundary value problems" (Celentano et al., 2024)
• "Talenti comparison results for solutions to $\omega_n$3-Laplace equation on multiply connected domains" (Barbato et al., 8 Apr 2025)
• "On a fractional boundary version of Talenti's inequality in the unit ball" (Karrouchi et al., 2024)
• "The sharp Poincaré--Sobolev type inequalities in the hyperbolic spaces $\omega_n$4" (Nguyen, 2018)
• "Isoperimetric inequalities and geometry of level curves of harmonic functions on smooth and singular surfaces" (Adamowicz et al., 2021)
• "Sharp quantitative Talenti's inequality in particular cases" (Acampora et al., 10 Mar 2025)
• "Sharp Sobolev inequalities via projection averages" (Kniefacz et al., 2019)
• "Sobolev and Hardy-Littlewood-Sobolev inequalities" (Dolbeault et al., 2013)
• "A note on the rearrangement of functions in time and on the parabolic Talenti inequality" (Idriss, 2022)