Eigenvalue Degeneracy and the Limits of Scalar Spectral Parametrisations on Compact Homogeneous Spaces

Spectral parametrizations in operator learning fall into two classes. One learns an arbitrary multiplier on the frequency lattice, as the Fourier neural operator does on the torus; the other learns a scalar function of the Laplace–Beltrami eigenvalue, as recent manifold constructions do. For spectral graph networks the second class is limited by the number of distinct eigenvalues, a limitation reported as rare on irregular graphs. On a compact homogeneous space, the situation reverses. Freudenthal’s formula confines the Casimir eigenvalues to a fixed lattice, so the number of distinct eigenvalues below Λ^2 is at most N_G Λ^2+1 whatever the dimension or rank, while the retained modes grow like Λ^n. On the d-torus the deficit is exact: order (logΛ)^(1/2) when d=2 by Landau’s theorem, a power of Λ when d≥3 by the three- and four-square theorems. The scalar class therefore omits the Fourier neural operator on every torus of dimension at least two. The bound persists for bundle-valued fields, where multiplicities enter quadratically; a numerical study separates capacity from approximation error, and the joint functional calculus of the invariant differential operators is identified as the constructive remedy.

spectral multiplier, Fourier multiplier, compact homogeneous space, Gelfand pair, eigenvalue multiplicity, neural operator, invariant differential operator.

1. Kovachki N, Li Z, Liu B, Azizzadenesheli K, Bhattacharya K, Stuart A, Anandkumar A. Neural operator: learning maps between function spaces with applications to PDEs. J Mach Learn Res. 2023;24(89):1–97. [Google Scholar] [Crossref]

2. Li Z, Kovachki N, Azizzadenesheli K, Liu B, Bhattacharya K, Stuart A, Anandkumar A. Fourier neural operator for parametric partial differential equations. In: Proceedings of the 9th International Conference on Learning Representations (ICLR); 2021. [Google Scholar] [Crossref]

3. Cheng P. Gauge-equivariant intrinsic neural operators for geometry-consistent learning of elliptic PDE maps. arXiv:2603.14734 [cs.LG]. 2026. [Google Scholar] [Crossref]

4. Tang H, Zhu J, Feng Z, Li H, Li C. Geometric Laplace neural operator. arXiv:2512.16409 [cs.LG]. 2025. [Google Scholar] [Crossref]

5. Wang X, Zhang M. How powerful are spectral graph neural networks. In: Proceedings of the 39th International Conference on Machine Learning (ICML). PMLR. 2022; 162:23341–23362. [Google Scholar] [Crossref]

6. Lu K, Yu Y, Fei H, Li X, Yang Z, Guo Z, Liang M, Yin M, Chua TS. Improving expressive power of spectral graph neural networks with eigenvalue correction. In: Proceedings of the 38th AAAI Conference on Artificial Intelligence. 2024;38(13):14158–14166. [Google Scholar] [Crossref]

7. Hordan S, Lifshitz G, Bechler-Speicher M, Dym N. Spectral graph neural networks are incomplete on graphs with a simple spectrum. In: Advances in Neural Information Processing Systems (NeurIPS). 2025. arXiv:2506.05530. [Google Scholar] [Crossref]

8. Helgason S. Groups and geometric analysis: integral geometry, invariant differential operators, and spherical functions. Orlando: Academic Press; 1984. [Google Scholar] [Crossref]

9. Ruzhansky M, Turunen V. Pseudo-differential operators and symmetries: background analysis and advanced topics. Pseudo-Differential Operators: Theory and Applications, vol. 2. Basel: Birkhäuser; 2010. [Google Scholar] [Crossref]

10. Chavel I. Eigenvalues in Riemannian geometry. Pure and Applied Mathematics, vol. 115. Orlando: Academic Press; 1984. [Google Scholar] [Crossref]

11. Bonev B, Kurth T, Hundt C, Pathak J, Baust M, Kashinath K, Anandkumar A. Spherical Fourier neural operators: learning stable dynamics on the sphere. In: Proceedings of the 40th International Conference on Machine Learning (ICML). PMLR. 2023;202:2806–2823. [Google Scholar] [Crossref]

12. Wallach NR. Harmonic analysis on homogeneous spaces. Pure and Applied Mathematics, vol. 19. New York: Marcel Dekker; 1973. [Google Scholar] [Crossref]

13. Zheng D, Zhong T, Allen-Blanchette C. Topology-preserving neural operator learning via Hodge decomposition. arXiv:2605.13834 [cs.LG]. 2026. [Google Scholar] [Crossref]

14. Kovachki N, Lanthaler S, Mishra S. On universal approximation and error bounds for Fourier neural operators. J Mach Learn Res. 2021;22(290):1–76. [Google Scholar] [Crossref]

15. Chandramoorthy N, Sanz-Alonso D, Waniorek N. From spectral methods to sample complexity bounds for Fourier neural operators. arXiv:2607.00320 [math.NA]. 2026. [Google Scholar] [Crossref]

• Interplay of Students’ Emotional Intelligence and Attitude toward Mathematics on Performance in Grade 10 Algebra
• Numerical Simulation of Fitzhugh-Nagumo Dynamics Using a Finite Difference-Based Method of Lines
• Fixed Point Theorem in Controlled Metric Spaces
• Usage of Moving Average to Heart Rate, Blood Pressure and Blood Sugar
• Exploring Algebraic Topology and Homotopy Theory: Methods, Empirical Data, and Numerical Examples