Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Nonnegative solutions to nonlocal parabolic equations

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read For nonlocal parabolic equations with bounded measurable coefficients, every nonnegative global weak solution is the convolution of a unique Radon initial measure with a fundamental solution, which is pointwise comparable to the…

desk verdict Genuinely new Widder-type and heat kernel estimates for nonlocal parabolic equations, but the central uniqueness proof in Proposition 5.4 has an unjustified WLOG assumption and a skipped time-mollification step that a referee should ask to be repaired. read the letter →

arxiv 2505.08449 v1 pith:BHMZ7Q5O submitted 2025-05-13 math.AP math.PR

Naian Liao , Marvin Weidner This is my paper · ORCID
classification math.APmath.PR MSC 47G2035B6535R0931B0535C1535K0835A02
keywords nonlocalparabolicequationsWidder-typetheoremfundamentalsolutionHarnackinequalitymeasurablecoefficientsinitialtracefractionalLaplacianintegro-differentialoperators
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

The paper aims to extend the classical representation theory of parabolic equations, where every nonnegative solution is a convolution of its initial data with a fundamental solution, to a broad class of nonlocal equations whose kernels are only bounded and measurable, not translation invariant. It proves that every nonnegative global weak solution in a natural weighted space has a unique Radon measure as its initial trace and is represented by integrating that measure against a fundamental solution. It then establishes that this fundamental solution is pointwise comparable to the fractional heat kernel, and derives a sharp Harnack-type estimate for all nonnegative global solutions. A sympathetic reader would care because these are structural facts: the whole solution theory of a large class of nonlocal diffusion equations is reduced to one kernel, and the proofs use only variational tools.

What carries the argument

The load-bearing object is the fundamental solution $p_t(x,y)$, the density that solves the Cauchy problem with a Dirac initial datum at $y$. The engine of the argument is the improved weak Harnack inequality (Theorem 3.3), which bounds the average of a nonnegative supersolution over a ball at early times by its infimum over a later cylinder; this inequality converts local information into the global weighted $L^1$ estimate (Lemma 5.1) that produces the initial trace and later forces uniqueness. The two-sided bounds of $p_t$ are obtained by combining the upper heat-kernel estimate from the previously developed analytic technique with a new lower estimate that uses the weak Harnack inequality with a tail term; this last step is purely variational and does not rely on semigroups, Dirichlet forms, or stochastic analysis.

What would settle it

Exhibit a kernel satisfying (1.5) and (1.6) for which the improved weak Harnack inequality (3.5) is false, for instance a time-dependent kernel engineered so that a nonnegative supersolution's average over $B_R$ at time $t_o$ decays faster than any constant multiple of its infimum over the later cylinder; since the paper imports this inequality rather than proving it, such a counterexample would invalidate Lemma 5.1 and the Widder-type theorem. A more targeted check is whether the vague limit of the averages $u(t,x)dx$ as $t\to0$ is unique for every nonnegative global solution in class (1.12): a single solution with two different subsequential limits would contradict Proposition 5.3.

Watch

Extended reading notes

Core claim

The central discovery is a nonlocal analogue of the classical Widder-type representation theorem. For any symmetric kernel satisfying the upper bound (1.5) and the integrated coercivity condition (1.6), every nonnegative global weak solution $u$ of $\partial_t u - L_t u = 0$ in the class $L^\infty_{\mathrm{loc}} L^2_{\mathrm{loc}} \cap L^2_{\mathrm{loc}} H^s_{\mathrm{loc}} \cap L^1_{\mathrm{loc}} L^1(\mathbb{R}^d;w)$ admits a unique nonnegative Radon measure $\mu$ with $\int d\mu/(1+|x|)^{d+2s} < \infty$, such that $u(t,x)=\int p_t(x,y)\,d\mu(y)$ and $u(0)=\mu$ in the vague sense; conversely, any such measure produces a solution by the same formula. Under the additional pointwise lower bound (1.7), the fundamental solution $p_t$ obeys the sharp two-sided estimate $c_1(t^{-d/2s} \wedge t/|x-y|^{d+2s}) \le p_t(x,y) \le c_2(t^{-d/2s} \wedge t/|x-y|^{d+2s})$, and every nonnegative global solution satisfies the sharp Harnack estimate $u(t,x)/u(\tau,y) \le c\,(t/\tau)^{-d/2s}(1+|x-y|/\tau^{1/2s})^{d+2s}$.

Load-bearing premise

The whole construction leans on an improved weak Harnack inequality, taken from the authors' previous work without proof: the average of a nonnegative supersolution over a ball at early times must be controlled by its infimum over a later cylinder, for every kernel satisfying only the upper bound (1.5) and the integrated coercivity condition (1.6); if that estimate fails, the weighted L1 bound, the initial trace, and hence the main representation theorem collapse.

Editorial extensions

If this is right

  • Every nonnegative global weak solution in the stated class is uniquely determined by its initial measure, so the Cauchy problem for nonlocal equations with measure data is well posed and the solution theory reduces to the study of the fundamental solution.
  • The fundamental solution of any operator whose kernel satisfies (1.5) and (1.7) is pointwise comparable to the fractional heat kernel up to constants depending only on $d$, $s$, $\lambda$, and $\Lambda$, including for time-dependent kernels.
  • The upper bound of the fundamental solution remains valid under the weaker integrated coercivity condition (1.6); only the matching lower bound requires the pointwise lower bound (1.7).
  • Every nonnegative global solution obeys the sharp Harnack estimate $u(t,x)/u(\tau,y) \le c\,(t/\tau)^{-d/2s}(1+|x-y|/\tau^{1/2s})^{d+2s}$ for $0<t\le\tau<T$, which is new even for the fractional heat equation.
  • Given the initial measure $\mu$, solutions grow at most like $u(t,x) \le c\,t^{-d/2s} \int ((t^{1/2s}+|x|)/(t^{1/2s}+|y|))^{d+2s}\,d\mu(y)$, quantifying how the initial datum controls later values.

Reading between the lines

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

  • Because the proof of the lower bound is purely variational, the same two-sided estimates are plausible for nonlinear nonlocal parabolic equations, where semigroup and stochastic tools are unavailable; testing this on fractional porous-medium-type equations would be a natural extension.
  • The paper leaves open whether the sharp Harnack estimate survives under the weaker coercivity condition (1.6); combining the Widder representation with lower heat-kernel bounds for such kernels would be a concrete test of that possibility.
  • Since the representation theorem identifies the initial trace without regularity up to $t=0$, it offers a canonical notion of initial data for nonlocal evolutions that may support studying singular limits such as vanishing fractional order or concentrating coefficients.
  • The growth estimate via initial data singles out the weighted measure space with norm $\int d\mu/(1+|x|)^{d+2s}$ as the natural initial-data space; a systematic study of whether condition (1.9) is optimal would test how sharp the theory is.
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 / 4 minor

Summary. The paper establishes a nonlocal analogue of Aronson's and Widder's theory for parabolic equations with bounded, measurable, symmetric kernels satisfying an upper bound and an integrated coercivity condition. The main results are a Widder-type representation theorem for nonnegative global weak solutions (Theorem 1.1), two-sided polynomial bounds for the fundamental solution (Theorem 1.3), a sharp elliptic-type Harnack estimate (Theorem 1.4), and a growth estimate in terms of the initial measure (Theorem 1.8). The proofs are variational and avoid semigroup, Dirichlet-form, and stochastic tools, with the construction of the fundamental solution and the lower heat-kernel bounds carried out directly from Harnack inequalities. The manuscript is generally well organized and carefully distinguishes the new contributions from the substantial inputs taken from the authors' earlier works.

Significance. If completed, the Widder-type theorem would be a significant advance: it extends a structural representation that was previously known only for translation-invariant operators to general nonlocal operators with measurable coefficients, and it identifies the correct growth class for the initial Radon measure. The two-sided fundamental solution bounds under time-dependent kernels and the short variational proof of the lower bound are also valuable, as is the sharp Harnack-type estimate for global solutions. The paper is transparent about its reliance on earlier results, especially the improved weak Harnack inequality from [LW24] and the upper bounds from [KW23]; this reliance is explicit rather than circular, though it means the central theorem is conditional on those black boxes. The main obstacle is that the uniqueness proof in Proposition 5.4 contains two under-specified steps that are load-bearing for Theorem 1.1(i), and the proof of Theorem 1.1 does not supply the missing arguments.

major comments (3)
  1. [Section 5.3, Proposition 5.4] The assertion 'Without loss of generality we assume \|u(T)\|_{L^1(R^d;w)} < \infty' is not justified. Definition 2.3 only guarantees u_1 and u_2 lie in L^1_loc((0,T); L^1(R^d;w)), so at the fixed terminal time T their weighted L^1 norms need not be finite or even well-defined as Lebesgue points. Moreover, u = u_1 - u_2 is signed, so Lemma 5.1 cannot be applied to u itself. The finiteness of \|u(T)\|_{L^1(R^d;w)} is used in the bound on III_1, in the estimate for III_2, and in the final dominated convergence as \tau_1 \nearrow T that yields \int \psi u(T) = 0. Without a proof that some terminal value of u lies in L^1(R^d;w), or a replacement of T by a Lebesgue point together with a separate continuity argument up to T, the uniqueness conclusion u(T)=0 is not established. This gap directly affects the central claim of Theorem 1.1(i).
  2. [Section 5.3, Proposition 5.4, time-mollification step] The display beginning 'Modulo a proper time mollification' differentiates the product \xi_R u \phi in time and then uses the weak formulations of u and \phi. The weak form only provides distributional time derivatives for u and \phi, so passing to the displayed identity requires a time mollification with uniform control of the resulting error terms as the mollification parameter is removed. This control is not supplied, and it is not immediate because the test functions are products involving the nonsmooth cutoff \xi_R and because the bilinear form E(t) has time-dependent kernel. Since this duality identity is exactly the mechanism that transports the terminal datum \psi back to \tau_o and then to the initial trace, the missing justification is load-bearing for the proof of uniqueness.
  3. [Section 5.5, proof of Theorem 1.1] The final step 'by Proposition 5.4, we deduce that u = v' applies Proposition 5.4 on the interval (0,T] and concludes equality at time T. However, Proposition 5.4's proof invokes the unjustified WLOG assumption on \|u(T)\|_{L^1(R^d;w)}, and Definition 2.3 defines solutions on (\eta,T) without prescribing regularity or values at the terminal time T. Thus the equality of the two solutions at time T is not actually obtained by the given arguments. The proof could be repaired by working with a terminal time T_o \in (0,T) that is a Lebesgue point for both solutions and then using local regularity to propagate equality, but this is not carried out in the manuscript.
minor comments (4)
  1. [Section 5.4, heading] The heading contains a typo: 'intitial data' should be 'initial data'.
  2. [Definition 2.3] The definition says 'Let [\eta,T] \subset R be an interval' but then refers to solutions on '(\eta+\epsilon,T)' and to convergence as t \searrow \eta. Please clarify whether the terminal time T is included in the solution interval and whether values such as u(T) are part of the solution concept; this is relevant to Proposition 5.4 and Theorem 1.1.
  3. [Lemma 5.1] The statement '\|u(t)\|_{L^1(R^d;w)} \le c\|u(T)\|_{L^1(R^d;w)}' is only meaningful when \|u(T)\|_{L^1(R^d;w)} < \infty. Please state explicitly that the estimate is asserted under this finiteness assumption, or clarify the convention when the right-hand side is infinite.
  4. [Remark 4.10] The sentence 'decomposing the space in the same way' is vague; please spell out the decomposition used to derive the lower bound in [Kan23] from (4.9), or give a precise reference to the relevant part of that paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; the paper's heavy self-citations are load-bearing but draw on prior independent results that do not contain the target theorems as assumptions.

full rationale

The central derivation chain is not circular. Theorem 1.1 is proved from Proposition 5.3 (initial trace via Lemma 5.1 and the improved weak Harnack inequality of [LW24]), Proposition 5.4 (uniqueness using the dual representation from Proposition 4.1, the upper bound (4.8), and Hölder regularity from [KW24]), and Proposition 5.5 (convolution with the fundamental solution). None of these imported results assumes the conclusion of Theorem 1.1. The improved weak Harnack inequality (3.5) is the engine of the proof, but it is a local estimate for nonnegative supersolutions, not a Widder-type representation for measure data. Similarly, the upper bound in Theorem 1.3 applies [KW23] after the present paper verifies the existence and symmetry assumptions for the fundamental solution; the lower bound relies on weak Harnack inequalities, not on the target two-sided estimate. Theorem 1.4 is derived from [LW24, Theorem 1.2 and Proposition 6.1], which are prior results with assumptions not including the sharp Harnack-type inequality being proved. The skeptic's concerns about 'Without loss of generality we assume ∥u(T)∥_{L^1(w)}<∞' and 'Modulo a proper time mollification' in Proposition 5.4 are proof-completeness issues, not circular reductions: they do not show that the conclusion has been assumed or fitted. No equation in the paper is used as both hypothesis and conclusion, and no fitted parameter is relabeled as a prediction. Self-citation is extensive and load-bearing, but under the stated hard rules that does not by itself constitute circularity.

Assumptions & free parameters 0 free parameters · 9 assumptions · 0 invented entities

No free parameters or invented entities are introduced. The central claims rest on the structural assumptions (1.5)-(1.7) on the kernel and on a substantial body of prior results, several from the authors' own earlier work ([LW24], [KW23], [KW24], [Lia24a], [Lia24b]), which are used as black boxes. The paper does not provide independent proofs of these inputs.

assumptions (9)
  • domain assumption Kernel upper bound (1.5): K(t;x,y) ≤ Λ |x-y|^{-d-2s}.
    Assumed in all theorems; controls large-distance interactions and is used in every estimate involving the tail or truncation (e.g., Lemmas 5.1, 4.9).
  • domain assumption Coercivity condition (1.6): local lower bound of the Dirichlet form by λ [v]^2_{H^s(B_r)}.
    Substantially weaker than a pointwise lower bound; underpins the weak Harnack inequalities (Theorems 3.1, 3.3) and the upper bound in Theorem 1.3.
  • domain assumption Pointwise lower bound (1.7): K(t;x,y) ≥ λ |x-y|^{-d-2s}.
    Required for the lower bound in Theorem 1.3 and for Theorem 1.4; the proof of Lemma 4.9 and the tail estimate (3.2) both use it. Without it the lower bound may fail (Remark 4.10).
  • standard math Improved weak Harnack inequality [LW24, Theorem 1.5], quoted as Theorem 3.3.
    Load-bearing for the weighted L^1 estimate (Lemma 5.1) and the initial trace (Proposition 5.3); it is cited, not reproved.
  • standard math Weak Harnack inequality with tail term [KW24, Theorem 1.9], quoted as Theorem 3.1.
    Used to prove the off-diagonal lower bound of the fundamental solution (Lemma 4.9).
  • standard math Existence, semigroup property, and symmetry of the L^2-Cauchy kernel p_{η,t} [LW24, Proposition A.1 / Proposition 4.1 here].
    Provides the kernel p_{η,t}, its positivity, Chapman-Kolmogorov identity (4.4), and symmetry (4.6), which are used throughout Section 4 and 5.
  • standard math Upper heat kernel estimate for nonlocal operators via Aronson's method [KW23, Theorem 2.3].
    Gives the off-diagonal upper bound (4.8) after verifying the existence and symmetry assumptions by this paper's construction.
  • standard math Hölder regularity estimates [KW24, Theorem 1.5; LW24, Theorem 3.8].
    Used in Appendix A (Liouville argument) and Appendix B (continuity up to initial time), and for the modulus estimate (5.4).
  • standard math Elliptic-type Harnack inequality and time-insensitive L^1(w) estimate [LW24, Theorem 1.2 and Proposition 6.1].
    Direct inputs in the proof of Theorem 1.4 and Theorem 1.8.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Nonnegative solutions to nonlocal parabolic equations." pith.science (2026). https://pith.science/paper/BHMZ7Q5O

@misc{pith2026250508449,
  author       = {Pith},
  title        = {Pith review of: Nonnegative solutions to nonlocal parabolic equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BHMZ7Q5O}},
  note         = {Machine review of arXiv:2505.08449}
}
read the original abstract

We aim to study nonnegative, global solutions to a general class of nonlocal parabolic equations with bounded measurable coefficients. First, we prove a Widder-type theorem. Such a result has previously been studied only for certain translation invariant operators, and new ideas are needed in our general setting. Second, we establish sharp two-sided bounds for the fundamental solution via purely variational techniques, entirely bypassing tools from semigroup theory, Dirichlet forms, and stochastic analysis. Third, we derive sharp Harnack-type estimates that are novel even for the fractional heat equation.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Liouville theorem for singular solutions to nonlocal equations

    math.AP 2025-07 conditional novelty 7.0 of 10

    Every singular solution to a nonlocal linear equation with measurable kernel that is one-sided bounded near zero and infinity must equal a multiple of the fundamental solution plus a constant.

Reference graph

Works this paper leans on

50 extracted references · 37 canonical work pages · cited by 1 Pith paper

  1. [1]

    H\"older regularity for fractional $p$-Laplace equations

    K. Adimurthi, H. Prasad, and V. Tewary. Local H \"o lder regularity for nonlocal parabolic p - L aplace equations. arXiv:2203.13082 (to appear in Ann. Sc. Norm. Super. Pisa Cl. Sci. (5)) , 2022

    work page Pith review arXiv 2022

  2. [2]

    D. G. Aronson. Non-negative solutions of linear parabolic equations. Ann. Scuola Norm. Sup. Pisa Cl. Sci. (3) , 22:607--694, 1968

    1968

  3. [3]

    D. G. Aronson. Addendum: `` N on-negative solutions of linear parabolic equations''\ ( A nn. S cuola N orm. S up. P isa (3) 22 (1968), 607--694). Ann. Scuola Norm. Sup. Pisa Cl. Sci. (3) , 25:221--228, 1971

    work page 1968

  4. [4]

    Barlow, A

    M. Barlow, A. Grigor'yan, and T. Kumagai. Heat kernel upper bounds for jump processes and the first exit time. J. Reine Angew. Math. , 626:135--157, 2009

    work page 2009

  5. [5]

    Barrios, I

    B. Barrios, I. Peral, F. Soria, and E. Valdinoci. A W idder's type theorem for the heat equation with nonlocal diffusion. Arch. Ration. Mech. Anal. , 213(2):629--650, 2014

    work page 2014

  6. [6]

    Bass and D

    R. Bass and D. Levin. Transition probabilities for symmetric jump processes. Trans. Amer. Math. Soc. , 354(7):2933--2953, 2002

    work page 2002

  7. [7]

    Blumenthal and R

    R. Blumenthal and R. Getoor. Some theorems on stable processes. Trans. Amer. Math. Soc. , 95:263--273, 1960

    work page 1960

  8. [8]

    Bonforte, Y

    M. Bonforte, Y. Sire, and J. V\' a zquez. Optimal existence and uniqueness theory for the fractional heat equation. Nonlinear Anal. , 153:142--168, 2017

    work page 2017

Show all 50 references
  1. [9]

    S.-S. Byun, H. Kim, and J. Ok. Local H \"older continuity for fractional nonlocal equations with general growth. Math. Ann. , 387(1-2):807--846, 2023

    2023

  2. [10]

    Caffarelli, C

    L. Caffarelli, C. H. Chan, and A. Vasseur. Regularity theory for parabolic nonlinear integral operators. J. Amer. Math. Soc. , 24(3):849--869, 2011

    2011

  3. [11]

    E. A. Carlen, S. Kusuoka, and D. W. Stroock. Upper bounds for symmetric M arkov transition functions. Ann. Inst. H. Poincar\'e Probab. Statist. , 23(2):245--287, 1987

    1987

  4. [12]

    Chaker, M

    J. Chaker, M. Kim, and M. Weidner. Regularity for nonlocal problems with non-standard growth. Calc. Var. Partial Differential Equations , 61(6):Paper No. 227, 31, 2022

    2022

  5. [13]

    Chaker, M

    J. Chaker, M. Kim, and M. Weidner. Harnack inequality for nonlocal problems with non-standard growth. Math. Ann. , 386(1-2):533--550, 2023

    2023

  6. [14]

    Z.-Q. Chen, P. Kim, T. Kumagai, and J. Wang. Heat kernel upper bounds for symmetric M arkov semigroups. J. Funct. Anal. , 281(4):Paper No. 109074, 40, 2021

    2021

  7. [15]

    Chen and T

    Z.-Q. Chen and T. Kumagai. Heat kernel estimates for stable-like processes on d -sets. Stochastic Process. Appl. , 108(1):27--62, 2003

    2003

  8. [16]

    Chen and T

    Z.-Q. Chen and T. Kumagai. Heat kernel estimates for jump processes of mixed types on metric measure spaces. Probab. Theory Related Fields , 140(1-2):277--317, 2008

    2008

  9. [17]

    Z.-Q. Chen, T. Kumagai, and J. Wang. Stability of heat kernel estimates for symmetric non-local D irichlet forms. Mem. Amer. Math. Soc. , 271(1330):v+89, 2021

    2021

  10. [18]

    Z.-Q. Chen, T. Kumagai, and J. Wang. Heat kernel estimates for general symmetric pure jump D irichlet forms. Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) , 23(3):1091--1140, 2022

    2022

  11. [19]

    M. Cozzi. Regularity results and H arnack inequalities for minimizers and solutions of nonlocal problems: a unified approach via fractional D e G iorgi classes. J. Funct. Anal. , 272(11):4762--4837, 2017

    2017

  12. [20]

    del Teso, J

    F. del Teso, J. Endal, and E. Jakobsen. Uniqueness and properties of distributional solutions of nonlocal equations of porous medium type. Adv. Math. , 305:78--143, 2017

    2017

  13. [21]

    Dembny and M

    M. Dembny and M. Sierzkega. A sharp H arnack bound for a nonlocal heat equation. arXiv:2303.08186 , 2023

    2023 arXiv

  14. [22]

    de Pablo, F

    A. de Pablo, F. Quir\'os, A. Rodr\'iguez, and J. L. V\'azquez. A general fractional porous medium equation. Comm. Pure Appl. Math. , 65(9):1242--1284, 2012

    2012

  15. [23]

    Di Castro, T

    A. Di Castro, T. Kuusi, and G. Palatucci. Nonlocal H arnack inequalities. J. Funct. Anal. , 267(6):1807--1836, 2014

    2014

  16. [24]

    Di Castro, T

    A. Di Castro, T. Kuusi, and G. Palatucci. Local behavior of fractional p -minimizers. Ann. Inst. H. Poincar\' e C Anal. Non Lin\' e aire , 33(5):1279--1299, 2016

    2016

  17. [25]

    Evans and R

    L. Evans and R. Gariepy. Measure theory and fine properties of functions . Textbooks in Mathematics. CRC Press, Boca Raton, FL, revised edition, 2015

    2015

  18. [26]

    Felsinger and M

    M. Felsinger and M. Kassmann. Local regularity for parabolic nonlocal operators. Comm. Partial Differential Equations , 38(9):1539--1573, 2013

    2013

  19. [27]

    Fern\'andez-Real and X

    X. Fern\'andez-Real and X. Ros-Oton. Regularity theory for general stable operators: parabolic equations. J. Funct. Anal. , 272(10):4165--4221, 2017

    2017

  20. [28]

    Grigor'yan, E

    A. Grigor'yan, E. Hu, and J. Hu. Lower estimates of heat kernels for non-local D irichlet forms on metric measure spaces. J. Funct. Anal. , 272(8):3311--3346, 2017

    2017

  21. [29]

    Grigor'yan, E

    A. Grigor'yan, E. Hu, and J. Hu. Two-sided estimates of heat kernels of jump type D irichlet forms. Adv. Math. , 330:433--515, 2018

    2018

  22. [30]

    Grigor'yan, E

    A. Grigor'yan, E. Hu, and J. Hu. Off-diagonal lower estimates and H \"older regularity of the heat kernel. Asian J. Math. , 27(5):675--770, 2023

    2023

  23. [31]

    Grigor'yan, E

    A. Grigor'yan, E. Hu, and J. Hu. Tail estimates and off-diagonal upper bounds of the heat kernel. preprint, https://www.math.uni-bielefeld.de/ grigor/tp-ueq.pdf , 2024

    2024

  24. [32]

    Grigor'yan, J

    A. Grigor'yan, J. Hu, and K.-S. Lau. Estimates of heat kernels for non-local regular D irichlet forms. Trans. Amer. Math. Soc. , 366(12):6397--6441, 2014

    2014

  25. [33]

    Grillo, M

    G. Grillo, M. Muratori, and F. Punzo. Fractional porous media equations: existence and uniqueness of weak solutions with measure data. Calc. Var. Partial Differential Equations , 54(3):3303--3335, 2015

    2015

  26. [34]

    Gonz \'a lvez, F

    I. Gonz \'a lvez, F. Quir \'o s, F. Soria, and Z. Vondracek. On the nonlocal heat equation for certain L \'e vy operators and the uniqueness of positive solutions. arXiv:2504.04246 , 2025

    2025 arXiv

  27. [35]

    J. Kang. Heat kernel estimates for symmetric jump processes with anisotropic jumping kernels. Proc. Amer. Math. Soc. , 151(1):385--399, 2023

    2023

  28. [36]

    Kassmann

    M. Kassmann. A priori estimates for integro-differential operators with measurable kernels. Calc. Var. Partial Differential Equations , 34(1):1--21, 2009

    2009

  29. [37]

    Kassmann and M

    M. Kassmann and M. Weidner. Upper heat kernel estimates for nonlocal operators via A ronson's method. Calc. Var. Partial Differential Equations , 62(2):Paper No. 68, 27, 2023

    2023

  30. [38]

    Kassmann and M

    M. Kassmann and M. Weidner. The parabolic H arnack inequality for nonlocal equations. Duke Math. J. , 173(17):3413--3451, 2024

    2024

  31. [39]

    Krzy\.za\'nski

    M. Krzy\.za\'nski. Sur les solutions non n\'egatives de l'\'equation lin\'eaire normale parabolique. Rev. Roumaine Math. Pures Appl. , 9:393--408, 1964

    1964

  32. [40]

    N. Liao. H arnack estimates for nonlocal drift-diffusion equations. arXiv:2402.11986v2 , 2024

    2024 arXiv

  33. [41]

    N. Liao. H\" o lder regularity for parabolic fractional p - L aplacian. Calc. Var. Partial Differential Equations , 63(1):Paper No. 22, 34 pp., 2024

    2024

  34. [42]

    N. Liao. On the modulus of continuity of solutions to nonlocal parabolic equations. J. Lond. Math. Soc. , 110(3):Paper No. e12985, 30 pp., 2024

    2024

  35. [43]

    Liao and M

    N. Liao and M. Weidner. Time-insensitive nonlocal parabolic H arnack estimates. arXiv:2409.20097 (to appear in Proc. Lond. Math. Soc.) , 2024

    2024 arXiv

  36. [44]

    Maekawa and H

    Y. Maekawa and H. Miura. On fundamental solutions for non-local parabolic equations with divergence free drift. Adv. Math. , 247:123--191, 2013

    2013

  37. [45]

    Maekawa and H

    Y. Maekawa and H. Miura. Upper bounds for fundamental solutions to non-local diffusion equations with divergence free drift. J. Funct. Anal. , 264(10):2245--2268, 2013

    2013

  38. [46]

    J. Moser. A H arnack inequality for parabolic differential equations. Comm. Pure Appl. Math. , 17:101--134, 1964

    1964

  39. [47]

    Ruiz-Cases

    J. Ruiz-Cases. Fractional fast diffusion with initial data a R adon measure. arXiv:2503.14296 , 2025

    2025 arXiv

  40. [48]

    Weber and R

    F. Weber and R. Zacher. Li- Y au inequalities for general non-local diffusion equations via reduction to the heat kernel. Math. Ann. , 385(1-2):393--419, 2023

    2023

  41. [49]

    M. Weidner. Energy methods for nonsymmetric nonlocal operators. PhD Thesis (Bielefeld University) , 2022

    2022

  42. [50]

    D. V. Widder. Positive temperatures on an infinite rod. Trans. Amer. Math. Soc. , 55:85--95, 1944

    1944

Pith tools

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