Peter–Weyl theorem

Peter–Weyl theorem
Name	Peter–Weyl theorem
Area	Representation theory
Introduced	1927
Authors	Fritz Peter; Hermann Weyl
Statement	Decomposition of square-integrable functions on compact groups into matrix coefficients of finite-dimensional representations

Peter–Weyl theorem

The Peter–Weyl theorem is a foundational result in the representation theory of compact topological groups, asserting that the matrix coefficients of irreducible finite-dimensional unitary representations form an orthonormal basis of L^2 functions on the group and yield a complete decomposition analogous to Fourier series. It connects harmonic analysis on compact groups with the theory of Fritz Peter, Hermann Weyl, David Hilbert's spectral ideas and the structure theory of Compact groups such as SU(2), SO(3), and toruses. The theorem underpins links between classical works by Élie Cartan, Hermann Minkowski, Évariste Galois-related symmetry ideas, and later developments by John von Neumann and Marcel Riesz.

Statement

Let G be a compact Hausdorff group equipped with a Haar measure μ. For each finite-dimensional continuous unitary representation ρ: G → U(n) and indices i,j, the matrix coefficient function g ↦ ρ(g)_{ij} belongs to L^2(G, μ). The theorem asserts that the linear span of all such matrix coefficients, taken over all irreducible unitary representations of G (up to equivalence), is dense in L^2(G, μ), and that distinct irreducible representations yield mutually orthogonal subspaces. Moreover, when normalized appropriately, the matrix coefficients form an orthonormal basis of L^2(G, μ), and the regular representation of G on L^2(G) decomposes as the Hilbert-space direct sum of the finite-dimensional irreducible representations, each occurring with multiplicity equal to its dimension. This formal statement synthesizes ideas from Peter and Weyl (1927) and connects to spectral decompositions used by John von Neumann.

Historical context and significance

The theorem originated in the 1920s work of Fritz Peter and Hermann Weyl as part of Weyl's program relating group representations to classical analysis and quantum theory evident in the works of Albert Einstein and Niels Bohr. It followed developments by Élie Cartan and Émile Borel on Lie groups and by Salomon Bochner on harmonic analysis. The Peter–Weyl result provided a rigorous analogue of the Fourier series expansion for noncommutative compact groups, influencing later contributions by Harish-Chandra, George Mackey, Israel Gelfand, Michael Atiyah, and Raoul Bott. Its significance extended to the representation theory of Lie groups such as SL(2,C), to quantum mechanics through Paul Dirac's formalism, and to topological methods employed by Hermann Weyl in his work on Invariance theory.

Proof outline

The proof combines analytic and algebraic techniques developed in the early 20th century. One begins by constructing the regular representation of G on L^2(G, μ) and uses compactness to show that convolution operators by continuous functions are compact and normal, invoking spectral theory from David Hilbert and Frigyes Riesz. Schur's lemma from Issai Schur and orthogonality relations for matrix coefficients provide algebraic orthogonality. A key step is approximating continuous functions by linear combinations of matrix coefficients via averaging arguments related to the Haar measure established by Alfréd Haar. The decomposition into finite-dimensional invariant subspaces is then obtained by showing those subspaces are spanned by irreducible unitary representations, completing the spectral decomposition in the spirit of John von Neumann's operator algebra frameworks.

Consequences and corollaries

Immediate consequences include the nonabelian generalization of the Fourier transform on compact groups, the Plancherel theorem for compact G, and explicit orthogonality relations for characters leading to character orthonormality used in the work of Frobenius and Burnside. The theorem implies that irreducible representations of compact groups are finite-dimensional, a fact exploited by Weyl in the classification of representations of compact Lie groups like SU(n), SO(n), and Sp(n). It yields the Peter–Weyl decomposition of C(G) into matrix blocks, feeding into index theory by Atiyah–Singer index theorem contributors such as Michael Atiyah and Isadore Singer, and informs the harmonic analysis techniques used in the Langlands program developed by Robert Langlands.

Examples and applications

For the circle group U(1), the theorem specializes to classical Fourier series with characters e^{inx}; this case relates to work by Joseph Fourier and later formalizations by Niels Henrik Abel. For SU(2), matrix coefficients are given by Wigner D-matrices used extensively in Eugene Wigner's studies of angular momentum in quantum mechanics and in molecular spectroscopy by Linus Pauling. In representation theory, Peter–Weyl underlies the decomposition of L^2 spaces on homogeneous spaces like S^n and informs computational techniques in numerical analysis used by researchers at institutions such as Courant Institute and Massachusetts Institute of Technology. The theorem also appears in modern quantum field theory contexts studied by groups at CERN and in noncommutative geometry advanced by Alain Connes.

Generalizations and extensions

Generalizations include the noncompact and nonunitary settings addressed by the Plancherel theorem for real reductive groups due to Harish-Chandra, the theory of unitary duals studied by George Mackey and Anthony Knapp, and extensions to locally compact groups with modular function considerations as treated by Jacques Dixmier and Paul Halmos. The role of matrix coefficients in ergodic theory connects to the work of Marian Rejewski and Yakov Sinai; categorical and operator-algebraic perspectives tie to Murray Gell-Mann? and to the development of C*-algebras by Gelfand–Naimark and von Neumann algebra theory by Murray and von Neumann. More recent research explores analogues in quantum groups developed by Drinfeld and Vladimir Drinfeld, and in p-adic groups related to the research program of Igor Frenkel and George Lusztig.

Category:Representation theory