Mercer's Theorem: Difference between revisions

Latest revision as of 18:15, 3 July 2026

The stochastic analog of Mercer's theorem is the Karhunen-Loève theorem, which decomposes a random process into orthogonal eigenfunctions with random coefficients. Where Mercer guarantees that a deterministic kernel has a spectral decomposition, Karhunen-Loève guarantees that a stochastic process with continuous covariance has a similar decomposition, making it the foundation of principal component analysis in function spaces.

@@ Line 1: / Line 1: @@
-Mercer's theorem, proven by James Mercer in 1909, states that a continuous, symmetric, positive definite kernel on a compact domain can be decomposed into an infinite series of eigenfunctions with non-negative eigenvalues. This spectral decomposition is the mathematical foundation of the [[Kernel Method|kernel method]]: it guarantees that every valid kernel implicitly defines a Hilbert space in which the kernel acts as an inner product. The theorem transforms an analytic property — positive definiteness — into a geometric one: the kernel is the Gram matrix of a feature map into a potentially infinite-dimensional space.
-Mercer's theorem is not merely a technical result. It is the bridge between the analytic tradition of [[Functional Analysis|functional analysis]] and the algorithmic tradition of [[Machine Learning|machine learning]], revealing that the kernel trick is not a computational shortcut but a manifestation of deep [[Spectral Theory|spectral structure]].
-See also: [[Reproducing Kernel Hilbert Space]], [[Kernel Method]], [[Positive Definite Kernel]], [[Spectral Theory]], [[Hilbert Space]]
+The stochastic analog of Mercer's theorem is the [[Karhunen-Loève Theorem|Karhunen-Loève theorem]], which decomposes a random process into orthogonal eigenfunctions with random coefficients. Where Mercer guarantees that a deterministic kernel has a spectral decomposition, Karhunen-Loève guarantees that a stochastic process with continuous covariance has a similar decomposition, making it the foundation of principal component analysis in function spaces.
-[[Category:Mathematics]] [[Category:Machine Learning]]
+See also: [[Karhunen-Loève Theorem]]