Pith. sign in

REVIEW 3 major objections 2 minor 113 references

Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels

T0 review · 3 major / 2 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 The paper establishes that the Wasserstein gradient flow of the squared MMD with the nonsmooth energy kernels $K(z)=-|z|^q$, $0<q<2$, is globally well-posed on $\mathbb{R}^d$ in subcritical $L^p$ classes of densities, and that the…

desk verdict First global well-posedness and particle limits for nonsmooth MMD energy flows in subcritical L^p, with an honest boundary and one imported endpoint; worth serious refereeing. read the letter →

arxiv 2608.01182 v1 pith:JBTEXQZI submitted 2026-08-02 math.AP math.PRstat.ML

Matthew Rosenzweig , Dejan Slepčev , Lihan Wang This is my paper · ORCID
classification math.APmath.PRstat.ML MSC 35Q7049Q2235B4035R0946E2260B1082C22
keywords maximummeandiscrepancyWassersteingradientflowenergykernelsRieszpotentialsaggregationequationmean-fieldlimitLagrangiancriticalpointnoncollision
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 tries to establish a complete dynamical theory for the Wasserstein gradient flow of the squared Maximum Mean Discrepancy built from the nonsmooth energy kernels $K(z)=-|z|^q$, $00$, global well-posedness nevertheless holds in $L^p$ spaces: unique weak solutions with Lipschitz velocity fields, an exact energy–dissipation identity, collision-free particle dynamics, a dimension-free $N^{-1/2}$ mean-field limit, and a near-complete rigidity classification of stationary states that identifies the target as the only absolutely continuous critical point outside two exceptional regimes. A sympathetic reader should care because the flow is the dynamical model behind MMD-based sampling and generative learning with these practical kernels, and because the paper pins down exactly what is and is not true about its convergence: no uniform rate exists, but qualitative approach to the target holds throughout most of the parameter range.

What carries the argument

The argument rests on four objects. (1) The singular-convolution bound (Lemma 3.2): for $0<a<d/p^*$, $\sup_x\int |x-y|^{-a}|f(y)|\,dy \le C\|f\|_{L^1}^{1-\theta}\|f\|_{L^p}^{\theta}$; applied with $a=2-q$ it makes convolutions of the kernel Hessian $|x|^{q-2}$ against $L^p$ densities bounded, yielding the Lipschitz velocity and driving the regularize–estimate–pass-to-the-limit–stability scheme for existence, plus the $W^1$ contraction for uniqueness. (2) The Fourier–Sobolev representation (Lemma 2.1) identifying $\mathrm{MMD}_q^2$ with the $\dot{H}^{-(d+q)/2}$ norm, which turns the energy into a Hilbert norm and the modulated energy into a coercive comparison functional. (3) The modulated-energy commutator estimate (Proposition 4.13) bounding the growth of the empirical-to-continuum MMD gap by velocity norms, giving the Gr\"onwall factor in the mean-field theorem. (4) The rigidity reduction: Lagrangian criticality plus Sobolev locality converts the critical-point equation into an equality of Riesz potentials on the positive part of the discrepancy, which the complete maximum principle for Riesz kernels and a one-sign cancellation argument close, leaving only the $d\in\{1,3\}$, $0<q<1$ exceptions.

What would settle it

Two concrete checks would settle open parts of the picture. (1) Well-posedness at the critical exponent $p=p_c=d/(d+q-2)$: proving uniqueness, or exhibiting non-uniqueness, there would fix the sharp threshold of Theorem 1.2. (2) Rigidity in the residual regime $d=3$, $0<q<1$: an absolutely continuous, non-radial probability density $\rho$ with finite $q$-th moment, unbounded positive-part discrepancy, and $\nabla K*(\rho-\mu)=0$ $\rho$-almost everywhere for some target $\mu$ would falsify the conjecture that natural-moment rigidity holds there. (3) A third check bears on target convergence: Corollary 5.25 rests on the uniform-in-time bound (5.95), so a solution with $1\le q<2$ whose $L^p$ norm or $r$-th moment grows without bound would separate asymptotic criticality from convergence to the target.

Watch

Extended reading notes

Core claim

On its own terms, the paper's claim is that the MMD Wasserstein flow of the energy kernel family is a well-posed evolution equation across its whole nondegenerate range. For $d+q-2>0$, any pair of probability densities with finite $r$-th moments ($r\ge 1$) and subcritical integrability $p\ge p_c=d/(d+q-2)$ (strictly supercritical when the bound is finite) generates a unique global weak solution $\rho_t$ of $\partial_t\rho_t = -\nabla\cdot(\rho_t v_t)$ with $v_t = -\nabla K*(\rho_t-\mu)$, the velocity lying in $L^\infty([0,T];W^{1,\infty})\cap C([0,T]\times\mathbb{R}^d)$, together with the energy–dissipation identity; the one-dimensional Coulomb endpoint $d=q=1$ is included via the companion Coulomb theory. The same framework yields: global noncollision and convergence to the critical set for the diagonal-free $N$-particle system; a modulated-energy estimate giving $N^{-1/2}$-rate mean-field convergence on finite intervals for well-prepared data; rigidity of Lagrangian critical points (an absolutely continuous critical state equals the target under finite $q$-moments, except possibly for $d\in\{1,3\}$ with $0<q<1$); and explicit obstructions — no initial-data-uniform MMD decay modulus and no global Polyak–\L{}ojasiewicz inequality on $\mathbb{R}^d$ or on $\mathbb{T}^d$ in the stated regimes.

Load-bearing premise

The entire theory depends on the source and target densities having finite moments and enough subcritical integrability ($p > d/(d+q-2)$) so that convolutions of the kernel's second derivative stay bounded; the critical borderline case $p=p_c$ is explicitly left open.

Editorial extensions

If this is right

  • For every kernel exponent $q$ with $d+q-2>0$ (and at the $d=q=1$ endpoint), the continuum flow is globally well-posed from any admissible density: the $r$-th moment stays bounded on finite intervals and the squared MMD equals its initial value minus the accumulated dissipation $\int |\nabla K*(\rho_\tau-\mu)|^2\,d\rho_\tau$.
  • The $N$-particle system with diagonal-free interactions has global collision-free solutions from any pairwise-distinct configuration; all particles remain bounded, their minimum separation is bounded away from zero uniformly in time, and the configuration converges to the collision-free critical set of the particle energy.
  • Well-prepared empirical measures (MMD error of order $N^{-1/2}$ for independent samples) stay within a Gr\"onwall factor of the continuum solution, giving convergence of the particle dynamics to the continuum flow as $N\to\infty$ on every finite time interval at rate $N^{-1/2}$.
  • Any absolutely continuous stationary state in the Lagrangian sense (the driving force vanishes on the carried mass) must equal the target under finite $q$-moments, except possibly in dimensions $1$ and $3$ for $0<q<1$; in the residual three-dimensional regime rigidity holds under an extra moment, compact support of the positive-part discrepancy, or radiality.
  • No initial-data-independent multiplicative MMD decay rate can hold on $\mathbb{R}^d$: translating a fixed compactly supported source rules out a single decay modulus, and global Polyak–\L{}ojasiewicz inequalities fail both in the whole space and in the periodic Riesz/Coulomb regimes stated.
  • Asymptotic criticality is unconditional for $1\le q<2$: every $\omega$-limit point of the continuum orbit is Lagrangian critical without additional long-time bounds, and the full orbit approaches the Lagrangian critical set; for $0<q<1$ the same holds under uniform moment and $L^p$ bounds, and rigidity then upgrades this to convergence to the target throughout the rigid part of the well-posedness

Reading between the lines

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

  • The paper leaves the critical integrability endpoint $p=p_c$ open and notes that velocity control degenerates from Lipschitz to Osgood there; the natural next step it gestures toward is an endpoint theory in a critical Lorentz class, and if uniqueness fails there, the sharp well-posedness threshold would be exactly the subcritical range proved here.
  • The rigidity exceptions ($d=1,3$ with $0<q<1$) and the explicit non-minimizing critical points suggest the long-time limit is genuinely selection-dependent in the flexible regimes: one could test numerically whether the basins of attraction of the non-minimizing critical states are nonempty for the continuum flow, which would show the dynamics, not the energy, decides the limit.
  • The finite-speed transport mechanism behind the no-rate obstruction implies that the practically meaningful object is the source's entry time into the target's bulk: the one-dimensional example shows post-entry relaxation is exponential when the target density is bounded below and polynomial when it vanishes, so rate estimates for these flows should be formulated after a datum-dependent waiting ti
  • The paper's saddle constructions and non-minimizing critical points suggest a wider principle: the empirical MMD energy landscape has collision-free critical states at all energy levels above the ground state, so deterministic particle trajectories can be trapped by symmetry; understanding which critical points admit recovery sequences of finite-particle equilibria would connect the static quantiz
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 / 2 minor

Summary. The paper studies the Wasserstein gradient flow of the squared maximum mean discrepancy generated by the energy kernels K(z)=-|z|^q, 0<q<2. It proves global well-posedness for probability densities in subcritical L^p classes when d+q-2>0, with a Lipschitz velocity field and an energy-dissipation identity, and it imports the one-dimensional Coulomb endpoint d=q=1 from the companion paper [CR26b]. It then proves global noncollision for the associated N-particle system, a fixed-N convergence-to-critical-set result, a modulated-energy mean-field estimate with explicit N^{-1/2}-type control for well-prepared data, a particle-to-continuum criticality principle, and saddle equilibria showing that deterministic trajectories need not reach global empirical minimizers. A large part of the paper is devoted to the stationary picture: Lagrangian critical points are shown to agree with the target in most regimes, with exceptions in d=1 and d=3 when 0<q<1, including explicit one-dimensional non-minimizing critical points. The paper also proves asymptotic criticality and conditional target convergence, and constructs explicit obstructions to any initial-data-independent MMD decay modulus and to global Polyak-Lojasiewicz inequalities. The critical integrability endpoint p=p_c for d+q-2>0 is explicitly left open (Remark 1.3, Section 9), and the mean-field estimate explicitly excludes d=1, 0<q<1 with nonzero target (Remark 1.9).

Significance. The results are substantial if the external inputs are accepted. The paper gives a detailed, parameter-free construction of global weak solutions in a regime where standard displacement-semiconvexity theory does not apply, together with quantitative particle-to-continuum propagation and a nearly complete classification of absolutely continuous Lagrangian critical points. The obstruction results are concrete and falsifiable, and the main theorem statements are carefully qualified: the strict subcritical condition p>p_c is stated, the critical endpoint is explicitly excluded, and the d=q=1 endpoint is delegated to [CR26b]. The four-step regularization argument, the virial-based noncollision proof, and the potential-theoretic rigidity arguments are presented in detail and are largely self-contained apart from the commutator estimate of [NRS22] and the companion endpoint inputs. The main correctness risk is the dependence of endpoint claims on self-cited preprints and a localized derivation error in the L^p estimate that appears to be fixable.

major comments (3)
  1. [§3.2, Eq. (3.36)] The displayed derivative of the L^p norm is incorrect as written. Multiplying (3.26) by p(ρ_ε)^{p-1} and integrating by parts gives d/dt ‖ρ_ε‖_p^p = (p-1)∫(ρ_ε)^p ΔK_ε*(ρ_ε-μ_ε) dx, not (p-1)∫(ρ_ε)^{p-1} ΔK_ε*(ρ_ε-μ_ε) dx. With the printed power, Hölder would only give a bound involving ‖ρ_ε‖_{L^{p-1}}^{p-1}, which is not controlled by ‖ρ_ε‖_{L^p}^p for probability densities with ‖ρ‖_{L^p}<1. The inequality in (3.37) is exactly what follows from the corrected formula, because ΔK_ε*(ρ_ε-μ_ε) ≤ (-ΔK_ε)*μ_ε ≤ ‖(-ΔK)*μ‖_{L^∞}. The four-step uniform-estimate construction therefore closes after this correction. Please correct (3.36) and re-verify the powers in (3.37)-(3.39) and (3.47)-(3.49), all of which rely on this estimate.
  2. [§3 (first paragraph), Remark 1.3, Proposition 4.13] The advertised one-dimensional Coulomb endpoint is not proved within this manuscript. Theorem 1.2 for d=q=1 is imported from [CR26b, Theorem 1.2 and Proposition 2.1], and the d+q≤2 case of the commutator estimate in Proposition 4.13, which is used in the endpoint case of Theorem 1.8, is delegated to [RS26, Theorem 4.1]. These are self-cited companion or preprint results whose status should be clarified. If the journal requires stated theorems to be supported by the manuscript or by published references, the authors should either include the endpoint arguments as an appendix or explicitly mark the d=q=1 claims in Theorems 1.2 and 1.8 as conditional on the acceptance of [CR26b] and [RS26]. The main d+q-2>0 finite-p theory appears self-contained apart from the published commutator estimate [NRS22].
  3. [Remark 1.3 and Section 9] The finite critical endpoint p=p_c for d+q-2>0 is left open, as the authors explicitly state. This is consistent with Theorem 1.2 as written, since condition (1.11) requires p>p_c whenever d+q-2>0. The paper should keep this qualification visible in the abstract and introduction; the current abstract's phrase 'subcritical L^p spaces' is accurate. The open problem is appropriately listed in Section 9 and does not by itself affect the stated theorems.
minor comments (2)
  1. [Proposition 4.13 / references] The citation [RS26, Theorem 4.1] is load-bearing in the d+q≤2 case of Proposition 4.13 but does not appear in the visible reference list in the provided text. Please ensure the full reference is included and that its publication status is indicated.
  2. [Abstract and Section 1] The abstract states that the one-dimensional Coulomb endpoint is 'included'; consider adding a parenthetical in the introduction that this endpoint is obtained from [CR26b], so that readers are not led to expect a fully self-contained proof of the endpoint case in this paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main claims are derived from stated hypotheses; companion citations supply independent endpoint and commutator results, not redefinitions of the conclusions.

full rationale

The central continuum result (Theorem 1.2) is proved self-containedly for d+q-2>0 from the regularized problem, the singular-convolution bound (Lemma 3.2), and the W1-stability estimate (Proposition 3.7). The subcritical condition p>p_c is a stated hypothesis, not an output of the proof, and the paper explicitly leaves the critical endpoint open: Remark 1.3 says 'The finite critical endpoint p=p_c<infty remains excluded; we return to the open problem of including it in Section 9.' The one-dimensional Coulomb endpoint is imported from [CR26b, Theorem 1.2 and Proposition 2.1], with Remark 1.3 stating plainly that 'this Coulomb endpoint follows instead from [CR26b]'; that is a citation to an independent companion proof, not a reduction of the present equations to their own inputs. The mean-field estimate uses the modulated-energy identity (Lemma 4.12), proved from the particle and continuum dynamics and the MMD representation, and then Grönwall's inequality; the commutator bounds invoked from [NRS22, Proposition 3.1] and [RS26, Theorem 4.1] are prior proved results with stated hypotheses, not consequences of the present theorem. The rigidity classification is obtained from Riesz-potential arguments (Propositions 5.8, 5.11, 5.14) and explicit equilibrium densities (Lemma 5.21, Proposition 5.22), rather than by renaming known results. No fitted parameter is renamed as a prediction, and no uniqueness theorem is invoked to force a choice: uniqueness in the strict range is proved in Proposition 3.7. Known limitations, including the excluded range in Remark 1.9 and the open critical-endpoint problem in Section 9, are disclosed explicitly, which further supports a non-circularity finding.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The central machinery rests on standard potential theory, standard ODE/PDE infrastructure, and two specialized external estimates from the authors' own prior work and from potential theory. There are no fitted parameters and no invented physical entities. The main caveat is the reliance on [NRS22, Proposition 3.1] and [CR26b] for load-bearing steps, and the stated exclusion of the critical L^{p_c} endpoint and of the one-dimensional 0 < q < 1 nonzero-target mean-field case.

assumptions (6)
  • standard math Schoenberg-Levy representation (2.2): |z|^q = c times integral of (1 - cos(2*pi*xi*z)) / |2*pi*xi|^(d+q) d xi for 0 < q < 2.
    Basis for identifying MMD_q with a negative homogeneous Sobolev norm in Lemma 2.1.
  • standard math Complete maximum principle for Riesz kernels, cited as [Zor23, Theorem 2.2].
    Used in Proposition 5.11 and Theorem 5.17 to convert finite Riesz contact into global potential inequalities and rigidity; not proved inside the paper.
  • domain assumption Commutator estimate of Nguyen-Rosenzweig-Serfaty, [NRS22, Proposition 3.1], for Riesz kernels with exponent in [d-2, d).
    Load-bearing in Proposition 4.13 for the transport commutator bound that drives the modulated-energy mean-field estimate. The paper gives an outline but relies on the cited theorem.
  • domain assumption Coulomb endpoint Cauchy theory of [CR26b, Theorem 1.2 and Proposition 2.1] for d = q = 1.
    The one-dimensional Coulomb endpoint in Theorem 1.2 is imported rather than reproved. This is an explicit external dependency on a companion preprint.
  • standard math Standard ODE and continuity-equation infrastructure: Picard-Lindelof, Peano's theorem, characteristic flows, and the AGS08 representation of solutions to the linear continuity equation.
    Used throughout Sections 3 and 4 for local existence of particle trajectories, flow representation, and uniqueness.
  • domain assumption Subcritical L^p integrability and finite moment hypotheses on source and target, as stated in Theorems 1.2, 1.5, and 1.8.
    All regularity and uniqueness estimates in Sections 3 and 4 rely on these hypotheses. The critical endpoint p = p_c is explicitly left open in Section 9, so the theorem is conditional on the stated integrability range.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels." pith.science (2026). https://pith.science/paper/JBTEXQZI

@misc{pith2026260801182,
  author       = {Pith},
  title        = {Pith review of: Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JBTEXQZI}},
  note         = {Machine review of arXiv:2608.01182}
}
abstract

We study the Wasserstein gradient flow of the squared Maximum Mean Discrepancy (MMD) generated by the nonsmooth energy kernels $K(z)=-|z|^q$, $0<q<2$. In dimensions $d\ge2$, the corresponding energies are not displacement semiconvex, so standard Wasserstein-gradient-flow theory does not apply. When $d+q-2>0$, we prove global well-posedness on $\mathbb{R}^d$ for probability densities in subcritical $L^p$ spaces, with targets in the same integrability class and with finite moments. We also include the one-dimensional Coulomb endpoint $d=q=1$. For the associated $N$-particle system, we prove global noncollision and fixed-$N$ convergence to the collision-free critical set, a particle-to-continuum criticality principle, and a modulated-energy mean-field estimate that yields convergence of the particle dynamics to the continuum flow as $N\to\infty$ on every finite time interval. We also construct collision-free saddle equilibria, showing that deterministic particle trajectories need not approach global empirical minimizers. For $1\le q<2$, every continuum solution in our class has a narrowly relatively compact orbit, every $\omega$-limit point is Lagrangian critical, and the orbit approaches the Lagrangian critical set. For $0<q<1$, the same conclusions hold under uniform-in-time moment and subcritical $L^p$ bounds. We prove that an absolutely continuous Lagrangian critical point equals the target when the source and target have finite moments of order $q$, except when $0<q<1$ and $d\in\{1,3\}$. Under the preceding uniform bounds, rigidity gives convergence of the continuum flow to the target throughout the rigid part of the well-posedness range. Finally, we show that no initial-data-independent multiplicative MMD decay modulus exists on $\mathbb{R}^d$, and that global Polyak--\L ojasiewicz inequalities fail in several whole-space and periodic Riesz/Coulomb regimes.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

113 extracted references · 52 canonical work pages

  1. [1]

    2026 , eprint=

    Quantitative Convergence of Wasserstein Gradient Flows of Kernel Mean Discrepancies , author=. 2026 , eprint=

    2026

  2. [2]

    Weighted minimum $\alpha$-Green energy problems

    Zorii, Natalia , title =. doi:10.48550/arXiv.2505.02260 , url =. 2505.02260 , eprintclass =

    work page Pith review arXiv doi:10.48550/arxiv.2505.02260

  3. [3]

    On the role of the point at infinity in Deny's principle of positivity of mass for Riesz potentials

    Zorii, Natalia , title =. Analysis and Mathematical Physics , volume =. 2023 , doi =. 2202.12418 , eprintclass =

    work page Pith review arXiv 2023

  4. [4]

    Riesz energy problems with external fields and related theory

    Dragnev, Peter and Orive, Ramon and Saff, Edward B. and Wielonsky, Franck , title =. Constructive Approximation , volume =. 2023 , doi =. 2104.03733 , eprintclass =

    work page Pith review arXiv 2023

  5. [5]

    Fractional Calculus and Applied Analysis , volume =

    Kwa\'snicki, Mateusz , title =. Fractional Calculus and Applied Analysis , volume =. doi:10.1515/fca-2017-0002 , eprinttype =. 1507.07356 , eprintclass =

    arXiv 2017

  6. [6]

    Wasserstein gradient flows for Coulomb discrepancies

    Chodron de Courcel, Antonin and Rosenzweig, Matthew , title =. doi:10.48550/arXiv.2607.12579 , url =. 2607.12579 , eprintclass =

    work page Pith review arXiv doi:10.48550/arxiv.2607.12579

  7. [7]

    Hess-Childs, Elias and Rosenzweig, Matthew and Serfaty, Sylvia , title =

  8. [8]

    Chodron de Courcel, Antonin and Rosenzweig, Matthew , title =

Show all 113 references
  1. [9]

    doi:10.1007/BFb0103945 , isbn =

    Graf, Siegfried and Luschgy, Harald , title =. doi:10.1007/BFb0103945 , isbn =

    doi

  2. [10]

    2605.18497 , archivePrefix =

    Colasanto, Francesco and Focardi, Matteo and Fornasier, Massimo and Mattesini, Francesco , title =. 2605.18497 , archivePrefix =

    arXiv

  3. [11]

    International Conference on Learning Representations , volume=

    Deep MMD gradient flow without adversarial training , author=. International Conference on Learning Representations , volume=. 2025 , eprint=

    2025

  4. [12]

    2015 , eprint=

    Absolute Moments and Fourier-Based Probability Metrics , author=. 2015 , eprint=

    2015

  5. [13]

    Li, Dong , journal=. On. 2019 , doi=. 1609.01780 , archivePrefix=

    2019 arXiv

  6. [14]

    ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=

    Generalized sliced probability metrics , author=. ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=. 2022 , organization=

    2022

  7. [15]

    Sejdinovic, Dino and Sriperumbudur, Bharath and Gretton, Arthur and Fukumizu, Kenji , TITLE =. Ann. Statist. , FJOURNAL =. 2013 , NUMBER =. doi:10.1214/13-AOS1140 , URL =

    2013 doi

  8. [16]

    Fournier, Nicolas and Guillin, Arnaud , TITLE =. Probab. Theory Related Fields , FJOURNAL =. 2015 , NUMBER =. doi:10.1007/s00440-014-0583-7 , URL =

    2015 doi

  9. [17]

    ESAIM Control Optim

    Duong, Richard and Stein, Viktor and Beinert, Robert and Hertrich, Johannes and Steidl, Gabriele , TITLE =. ESAIM Control Optim. Calc. Var. , FJOURNAL =. 2026 , PAGES =. doi:10.1051/cocv/2025097 , URL =

    2026

  10. [18]

    Conference on learning theory , pages=

    Mean-field theory of two-layers neural networks: dimension-free bounds and kernel limit , author=. Conference on learning theory , pages=. 2019 , organization=

    2019

  11. [19]

    Communications on Pure and Applied Mathematics , volume=

    Trainability and Accuracy of Artificial Neural Networks: An Interacting Particle System Approach , author=. Communications on Pure and Applied Mathematics , volume=. 2022 , doi=. 1805.00915 , archivePrefix=

    2022 arXiv

  12. [20]

    Stochastic Processes and their Applications , volume=

    Mean field analysis of neural networks: A central limit theorem , author=. Stochastic Processes and their Applications , volume=. 2020 , doi=

    2020

  13. [21]

    NeurIPS , year=

    On the global convergence of gradient descent for over-parameterized models using optimal transport , author=. NeurIPS , year=

  14. [22]

    Proceedings of the 41st International Conference on Machine Learning , series=

    Mean-field Analysis on Two-layer Neural Networks from a Kernel Perspective , author=. Proceedings of the 41st International Conference on Machine Learning , series=. 2024 , publisher=. 2403.14917 , archivePrefix=

    2024 arXiv

  15. [23]

    Lopes, Orlando , journal=

  16. [24]

    4: applications of harmonic analysis , author=

    Generalized functions, Vol. 4: applications of harmonic analysis , author=. 1964 , publisher=

    1964

  17. [25]

    2019 , publisher=

    Probability: theory and examples , author=. 2019 , publisher=

    2019

  18. [26]

    Equivalence of gradient flows and entropy solutions for singular nonlocal interaction equations in

    Bonaschi, Giovanni A and Carrillo, Jos. Equivalence of gradient flows and entropy solutions for singular nonlocal interaction equations in. ESAIM: Control, Optimisation and Calculus of Variations , volume=. 2015 , doi=

    2015

  19. [27]

    Wasserstein steepest descent flows of discrepancies with

    Hertrich, Johannes and Gr. Wasserstein steepest descent flows of discrepancies with. Journal of Mathematical Analysis and Applications , volume=. 2024 , publisher=

    2024

  20. [28]

    Generalization of an inequality by

    Otto, Felix and Villani, C. Generalization of an inequality by. Journal of Functional Analysis , volume=. 2000 , publisher=

    2000

  21. [29]

    2008 , publisher=

    Gradient flows in metric spaces and in the space of probability measures , author=. 2008 , publisher=

    2008

  22. [30]

    Communications on Pure and Applied Mathematics , volume=

    Bertozzi, Andrea L and Laurent, Thomas and Rosado, Jes. Communications on Pure and Applied Mathematics , volume=. 2011 , publisher=

    2011

  23. [31]

    Communications in Partial Differential Equations , volume=

    Local and global existence for an aggregation equation , author=. Communications in Partial Differential Equations , volume=. 2007 , publisher=

    2007

  24. [32]

    European Congress of Mathematics , volume=

    Uniqueness of bounded solutions to aggregation equations by optimal transport methods , author=. European Congress of Mathematics , volume=

  25. [33]

    Sharp conditions to avoid collisions in singular

    Carrillo, Jos. Sharp conditions to avoid collisions in singular. Nonlinear Analysis: Real World Applications , volume=. 2017 , doi=

    2017

  26. [34]

    Uniqueness of the solution to the

    Loeper, Gr. Uniqueness of the solution to the. Journal de math. 2006 , publisher=

    2006

  27. [35]

    Communications on pure and applied mathematics , volume=

    Polar factorization and monotone rearrangement of vector-valued functions , author=. Communications on pure and applied mathematics , volume=. 1991 , publisher=

    1991

  28. [36]

    A computational fluid mechanics solution to the

    Benamou, Jean-David and Brenier, Yann , journal=. A computational fluid mechanics solution to the. 2000 , publisher=

    2000

  29. [37]

    Acta Mathematica , volume=

    The geometry of optimal transportation , author=. Acta Mathematica , volume=. 1996 , publisher=

    1996

  30. [38]

    Advances in mathematics , volume=

    A convexity principle for interacting gases , author=. Advances in mathematics , volume=. 1997 , publisher=

    1997

  31. [39]

    Finite-time blow-up of

    Bertozzi, Andrea L and Brandman, Jeremy , journal=. Finite-time blow-up of

  32. [40]

    Classical

    Grafakos, Loukas , journal=. Classical. 2008 , publisher=

    2008

  33. [41]

    The Journal of Machine Learning Research , volume=

    A kernel two-sample test , author=. The Journal of Machine Learning Research , volume=. 2012 , publisher=

    2012

  34. [42]

    On the global convergence of

    Boufad. On the global convergence of. SIAM Journal on Mathematical Analysis , volume=. 2025 , doi=. 2312.00800 , archivePrefix=

    2025

  35. [43]

    Journal of Differential Equations , volume=

    An invariance principle for gradient flows in the space of probability measures , author=. Journal of Differential Equations , volume=. 2023 , doi=

    2023

  36. [44]

    Characterization of translation invariant

    Modeste, Thibault and Dombry, Cl. Characterization of translation invariant. Journal of Machine Learning Research , volume=

  37. [45]

    2021 , publisher=

    Topics in optimal transportation , author=. 2021 , publisher=

    2021

  38. [46]

    SIAM Journal on Mathematical Analysis , author=

    Mean-. SIAM Journal on Mathematical Analysis , author=. 2016 , pages=. doi:10.1137/15M1042620 , number=

    2016 doi

  39. [47]

    Lectures on

    Serfaty, Sylvia , series=. Lectures on. 2026 , pagetotal=. doi:10.1090/coll/070 , eprinttype=. 2407.21194 , eprintclass=

    2026 arXiv

  40. [48]

    Sharp commutator estimates of all order for

    Rosenzweig, Matthew and Serfaty, Sylvia , journal=. Sharp commutator estimates of all order for. 2026 , doi=

    2026

  41. [49]

    Mean field limit for

    Serfaty, Sylvia , journal=. Mean field limit for. 2020 , publisher=

    2020

  42. [50]

    Grafakos, Loukas , volume=. Modern. 2009 , publisher=

    2009

  43. [51]

    Mean-field limits of

    Nguyen, Quoc Hung and Rosenzweig, Matthew and Serfaty, Sylvia , journal=. Mean-field limits of. 2022 , number=. doi:10.15781/nvv7-jy87 , eprint=

    2022 doi

  44. [52]

    Hitchhiker's guide to the fractional

    di Nezza, Eleonora and Palatucci, Giampiero and Valdinoci, Enrico , journal=. Hitchhiker's guide to the fractional. 2012 , publisher=

    2012

  45. [53]

    Sharp constants in the

    Lieb, Elliott H , booktitle=. Sharp constants in the. 1983 , publisher=

    1983

  46. [54]

    2001 , publisher=

    Analysis , author=. 2001 , publisher=

    2001

  47. [55]

    On the Best Values of the Constants in the Theorems of

    Pichorides, Stylianus Constantine , year=. On the Best Values of the Constants in the Theorems of

  48. [56]

    , author=

    Riesz transforms and related singular integrals. , author=. Journal f. 1996 , publisher=

    1996

  49. [57]

    1970 , publisher=

    Singular Integrals and Differentiability Properties of Functions , author=. 1970 , publisher=

    1970

  50. [58]

    Sharp inequalities for martingales with applications to the

    Ba. Sharp inequalities for martingales with applications to the. Duke Mathematical Journal , volume=. 1995 , publisher=

    1995

  51. [59]

    The variational formulation of the

    Jordan, Richard and Kinderlehrer, David and Otto, Felix , journal=. The variational formulation of the. 1998 , publisher=

    1998

  52. [60]

    Advances in neural information processing systems , volume=

    Generative modeling by estimating gradients of the data distribution , author=. Advances in neural information processing systems , volume=

  53. [61]

    Advances in neural information processing systems , volume=

    Improved techniques for training score-based generative models , author=. Advances in neural information processing systems , volume=

  54. [62]

    Score-Based Generative Modeling through Stochastic Differential Equations , author=

  55. [63]

    Advances in neural information processing systems , volume=

    Denoising diffusion probabilistic models , author=. Advances in neural information processing systems , volume=

  56. [64]

    Sampling as optimization in the space of measures: The

    Wibisono, Andre , booktitle=. Sampling as optimization in the space of measures: The. 2018 , organization=

    2018

  57. [65]

    Liu, Qiang and Wang, Dilin , journal=

  58. [66]

    Accelerating

    Lu, Yulong and Lu, Jianfeng and Nolen, James , journal=. Accelerating

  59. [67]

    Variational inference via

    Lambert, Marc and Chewi, Sinho and Bach, Francis and Bonnabel, Silv. Variational inference via. Advances in Neural Information Processing Systems , volume=

  60. [68]

    Advances in Neural Information Processing Systems , volume=

    Maximum mean discrepancy gradient flow , author=. Advances in Neural Information Processing Systems , volume=

  61. [69]

    Korba, Anna and Aubin-Frankowski, Pierre-Cyril and Majewski, Szymon and Ablin, Pierre , booktitle=. Kernel. 2021 , organization=

    2021

  62. [70]

    Interacting

    Garbuno-Inigo, Alfredo and Hoffmann, Franca and Li, Wuchen and Stuart, Andrew M , journal=. Interacting. 2020 , publisher=

    2020

  63. [71]

    Communications in applied mathematics and computational science , volume=

    Ensemble samplers with affine invariance , author=. Communications in applied mathematics and computational science , volume=. 2010 , publisher=

    2010

  64. [72]

    Nonlinearity , volume=

    Birth-death dynamics for sampling: global convergence, approximations and their asymptotics , author=. Nonlinearity , volume=. 2023 , publisher=

    2023

  65. [73]

    Ensemble

    Lindsey, Michael and Weare, Jonathan and Zhang, Anna , journal=. Ensemble. 2022 , publisher=

    2022

  66. [74]

    Studies in Applied Mathematics , volume=

    Consensus-based sampling , author=. Studies in Applied Mathematics , volume=. 2022 , publisher=

    2022

  67. [75]

    Communications in Mathematical Sciences , volume=

    Second order ensemble Langevin method for sampling and inverse problems , author=. Communications in Mathematical Sciences , volume=. 2025 , publisher=

    2025

  68. [76]

    Advances in applied probability , volume=

    Integral probability metrics and their generating classes of functions , author=. Advances in applied probability , volume=. 1997 , publisher=

    1997

  69. [77]

    Regularized

    He, Ye and Balasubramanian, Krishnakumar and Sriperumbudur, Bharath K and Lu, Jianfeng , journal=. Regularized. 2024 , publisher=

    2024

  70. [78]

    Liutkus, Antoine and Simsekli, Umut and Majewski, Szymon and Durmus, Alain and St. Sliced-. International Conference on machine learning , pages=. 2019 , organization=

    2019

  71. [79]

    Proceedings of the National Conference on Artificial Intelligence , volume=

    A kernel approach to comparing distributions , author=. Proceedings of the National Conference on Artificial Intelligence , volume=. 2007 , publisher=

    2007

  72. [80]

    Journal of Machine Learning Research , volume=

    Minimax estimation of kernel mean embeddings , author=. Journal of Machine Learning Research , volume=

  73. [81]

    Bernoulli , volume=

    On the optimal estimation of probability measures in weak and strong topologies , author=. Bernoulli , volume=

  74. [82]

    Universality, Characteristic Kernels and

    Sriperumbudur, Bharath K and Fukumizu, Kenji and Lanckriet, Gert RG , journal=. Universality, Characteristic Kernels and

  75. [83]

    2008 , publisher=

    Support vector machines , author=. 2008 , publisher=

    2008

  76. [84]

    (De)-regularized

    Chen, Zonghao and Mustafi, Aratrika and Glaser, Pierre and Korba, Anna and Gretton, Arthur and Sriperumbudur, Bharath K , journal=. (De)-regularized. 2025 , eprint=

    2025

  77. [85]

    Physica D: Nonlinear Phenomena , volume=

    Nonlocal interactions by repulsive--attractive potentials: radial ins/stability , author=. Physica D: Nonlinear Phenomena , volume=. 2013 , publisher=

    2013

  78. [86]

    Altekr. Neural. Proceedings of the 40th International Conference on Machine Learning , pages=

  79. [87]

    Posterior Sampling Based on Gradient Flows of the

    Hagemann, Paul and Hertrich, Johannes and Altekr. Posterior Sampling Based on Gradient Flows of the

  80. [88]

    Generative Sliced

    Hertrich, Johannes and Wald, Christian and Altekr. Generative Sliced. The Twelfth International Conference on Learning Representations , year=

  81. [89]

    Smoothed distance kernels for

    Rux, Nicolaj and Quellmalz, Michael and Steidl, Gabriele , journal=. Smoothed distance kernels for. 2026 , doi=

    2026

  82. [90]

    Journal of Multivariate Analysis , volume=

    A new test for multivariate normality , author=. Journal of Multivariate Analysis , volume=. 2005 , publisher=

    2005

  83. [91]

    Bowling Green State University, Department of Mathematics and Statistics Technical Report , volume=

    E-statistics: The energy of statistical samples , author=. Bowling Green State University, Department of Mathematics and Statistics Technical Report , volume=

  84. [92]

    Journal of statistical planning and inference , volume=

    Energy statistics: A class of statistics based on distances , author=. Journal of statistical planning and inference , volume=. 2013 , publisher=

    2013

  85. [93]

    Journal of Machine Learning Research , volume=

    Metrizing weak convergence with maximum mean discrepancies , author=. Journal of Machine Learning Research , volume=

  86. [94]

    Global-in-time mean-field convergence for singular

    Rosenzweig, Matthew and Serfaty, Sylvia , journal=. Global-in-time mean-field convergence for singular. 2023 , publisher=

    2023

  87. [95]

    Commutators, mean-field, and supercritical mean-field limits for

    Rosenzweig, Matthew , journal=. Commutators, mean-field, and supercritical mean-field limits for. 2025 , doi=

    2025

  88. [96]

    Wasserstein distances for vortices approximation of

    Hauray, Maxime , journal=. Wasserstein distances for vortices approximation of. 2009 , doi=

    2009

  89. [97]

    Ecole d'

    Topics in propagation of chaos , author=. Ecole d'. 1991 , publisher=

    1991

  90. [98]

    Quantitative estimates of propagation of chaos for stochastic systems with

    Jabin, Pierre-Emmanuel and Wang, Zhenfu , journal=. Quantitative estimates of propagation of chaos for stochastic systems with. 2018 , publisher=

    2018

  91. [99]

    Active Particles, Volume 4: Theory, Models, Applications , pages=

    Bayesian sampling using interacting particles , author=. Active Particles, Volume 4: Theory, Models, Applications , pages=. 2024 , publisher=

    2024

  92. [100]

    Hardin, D. P. and Reznikov, A. and Saff, E. B. and Volberg, A. , journal=. Local Properties of. 2019 , doi=

    2019

  93. [101]

    Functional Analysis and Its Applications , volume=

    Vlasov equations , author=. Functional Analysis and Its Applications , volume=. 1979 , doi=

    1979

  94. [102]

    2019 , doi=

    Discrete Energy on Rectifiable Sets , author=. 2019 , doi=

    2019

  95. [103]

    Analysis and Mathematical Physics , volume=

    Riesz Capacity: Monotonicity, Continuity, Diameter and Volume , author=. Analysis and Mathematical Physics , volume=. 2025 , doi=

    2025

  96. [104]

    and Rodiac, R

    Ambrosio, Luigi and Ponce, Augusto C. and Rodiac, R. Critical weak-. Rev. Mat. Iberoam. , volume=. 2020 , doi=

    2020

  97. [105]

    2015 , isbn=

    Measure Theory and Fine Properties of Functions , author=. 2015 , isbn=

    2015

  98. [106]

    Archive for Rational Mechanics and Analysis , volume=

    Nonlinear porous medium flow with fractional potential pressure , author=. Archive for Rational Mechanics and Analysis , volume=. 2011 , doi=

    2011

  99. [107]

    Journal of the European Mathematical Society , volume=

    Regularity of solutions of the fractional porous medium flow , author=. Journal of the European Mathematical Society , volume=. 2013 , doi=

    2013

  100. [108]

    The nonlocal porous medium equation:

    Biler, Piotr and Imbert, Cyril and Karch, Grzegorz , journal=. The nonlocal porous medium equation:. 2015 , doi=

    2015

  101. [109]

    Nonlinear Analysis , volume=

    Classical solutions for fractional porous medium flow , author=. Nonlinear Analysis , volume=. 2021 , doi=

    2021

  102. [110]

    2012 , doi=

    Ordinary Differential Equations and Dynamical Systems , author=. 2012 , doi=

    2012

  103. [111]

    Acta Mathematica , volume=

    On the Theory of Potentials in Locally Compact Spaces , author=. Acta Mathematica , volume=. 1960 , doi=

    1960

  104. [112]

    1972 , isbn=

    Foundations of Modern Potential Theory , author=. 1972 , isbn=

    1972

  105. [113]

    Annales de l'Institut Fourier , volume=

    Kurdyka, Krzysztof , title=. Annales de l'Institut Fourier , volume=. 1998 , doi=

    1998

Pith tools

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