Invariant kernels on Riemannian symmetric spaces: a harmonic-analytic approach
Not on the map. Why?
Abstract (arXiv)
This work aims to prove that the classical Gaussian kernel, when defined on a non-Euclidean symmetric space, is never positive-definite for any choice of parameter. To achieve this goal, the paper develops new geometric and analytical arguments. These provide a rigorous characterization of the positive-definiteness of the Gaussian kernel, which is complete but for a limited number of scenarios in low dimensions that are treated by numerical computations. Chief among these results are the L!scriptscriptstyle p-hspace0.02cmGodement theorems (where p = 1,2), which provide verifiable necessary and sufficient conditions for a kernel defined on a symmetric space of non-compact type to be positive-definite. A celebrated theorem, sometimes called the Bochner-Godement theorem, already gives such conditions and is far more general in its scope, but is especially hard to apply. Beyond the connection with the Gaussian kernel, the new results in this work lay out a blueprint for the study of invariant kernels on symmetric spaces, bringing forth specific harmonic analysis tools that suggest many future applications.
Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 10 of its 56 references that are in the map. First posted on arXiv 30 Oct 2023.
Builds on
10 of 56 references.
References · 10
- 2020Mat'ern Gaussian processes on Riemannian manifoldscited by 82
- 2014Geodesic Exponential Kernels: When Curvature and Linearity Conflictcited by 47
- 2014Kernel Methods on Riemannian Manifolds with Gaussian RBF Kernelscited by 30
- 2022Stationary Kernels and Gaussian Processes on Lie Groups and their Homogeneous Spaces I: the compact casecited by 25
- 2021Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approachcited by 23
- 2023Stationary Kernels and Gaussian Processes on Lie Groups and their Homogeneous Spaces II: non-compact symmetric spacescited by 15
- 2023The Gaussian kernel on the circle and spaces that admit isometric embeddings of the circlecited by 5
- 2023Gaussian kernels on non-simply-connected closed Riemannian manifolds are never positive definitecited by 3
- 2023Geometric Learning with Positively Decomposable Kernelscited by 3
- 2022Geometric Learning of Hidden Markov Models via a Method of Moments Algorithmcited by 2
Built on by
Citing papers per year: 2023: 1; 2024: 1; 2025: 2; 2026: 2.
6 papers.
- 2023Sample Path Regularity of Gaussian Processes from the Covariance Kernelcited by 11
- 2024The GeometricKernels Package: Heat and Mat'ern Kernels for Geometric Learning on Manifolds, Meshes, and Graphscited by 10
- 2025Memory Capacity of Nonlinear Recurrent Networks: Is it Informative?cited by 3
- 2025Universal kernels via harmonic analysis on Riemannian symmetric spacescited by 2
- 2026Heat and Mat'ern Kernels on Matchingscited by 0
- 2026RW-Flow: One-Step Generation on Compact Manifolds via Wasserstein Gradient Flowscited by 0
- 2026RW-Flow: One-Step Generation on Compact Manifolds via Wasserstein Gradient Flows30 Sep 2026
- 2026Heat and Mat'ern Kernels on Matchings15 Apr 2026
- 2025Universal kernels via harmonic analysis on Riemannian symmetric spaces24 Jun 2025
- 2025Memory Capacity of Nonlinear Recurrent Networks: Is it Informative?7 Feb 2025
- 2024The GeometricKernels Package: Heat and Mat'ern Kernels for Geometric Learning on Manifolds, Meshes, and Graphs10 Jul 2024
- 2023Sample Path Regularity of Gaussian Processes from the Covariance Kernel22 Dec 2023