| Inner product space | |
|---|---|
| Name | Inner product space |
| Type | Algebraic structure |
| Axioms | Positive-definiteness; (sesqui)linearity; conjugate symmetry |
| Related | Hilbert space; Banach space; Linear operator; Spectral theorem |
Inner product space An inner product space is a vector space equipped with an inner product: a positive-definite, sesquilinear (or bilinear) form that induces a norm and notions of angle and orthogonality. In quantum physics, inner product spaces provide the mathematical setting for state vectors, probability amplitudes, and the geometry underlying the superposition principle, measurement, and unitary evolution.
An inner product space is a pair (V, ⟨·,·⟩) where V is a vector space over the field ℂ or ℝ and ⟨·,·⟩: V × V → ℂ (or ℝ) satisfies: - Conjugate symmetry: ⟨u,v⟩ = overline{⟨v,u⟩}. - Linearity in the first (or second) slot and conjugate linearity in the other (sesquilinearity) for complex spaces, or bilinearity for real spaces. - Positive-definiteness: ⟨v,v⟩ ≥ 0 with equality iff v = 0.
From the inner product one defines the norm ||v|| = sqrt(⟨v,v⟩), the notion of orthogonality (⟨u,v⟩ = 0), and the induced metric. Fundamental inequalities and identities include the Cauchy–Schwarz inequality, the triangle inequality, and the polarization identity which reconstructs the inner product from the norm. These properties underpin many constructions in functional analysis and the mathematical formalism of Quantum mechanics.
Standard finite-dimensional example: the Euclidean inner product on ℝ^n or ℂ^n, ⟨x,y⟩ = Σ_i x_i overline{y_i}, used in linear algebra and matrix theory (e.g., Hermitian matrix properties). Function-space examples include L^2 spaces: for a measure space (X, μ), the inner product on L^2(X, μ) is ⟨f,g⟩ = ∫_X f(x) overline{g(x)} dμ(x); these are central in quantum theory as wavefunction spaces over configuration space. Sequence spaces: the ℓ^2 space of square-summable sequences with ⟨a,b⟩ = Σ_n a_n overline{b_n} models countable orthonormal bases, such as occupation-number bases in Fock space constructions.
Named constructions and examples appear in the literature of John von Neumann and David Hilbert; many canonical orthonormal systems (Fourier basis, Hermite functions) are studied in the context of harmonic analysis and operator theory at institutions such as Princeton University and University of Göttingen where foundational work in functional analysis took place.
An inner product space that is complete with respect to the norm topology is a Hilbert space. Completeness guarantees limits of Cauchy sequences exist and allows spectral decomposition, orthonormal basis expansions, and the Riesz representation theorem. Hilbert spaces such as L^2(ℝ^3) or separable spaces ℓ^2 are the canonical state spaces in nonrelativistic quantum mechanics and quantum field theory. Important results—Riesz representation, Parseval's identity, and orthonormal basis existence—are used in treatments by authors like Reed and Simon and texts from Cambridge University Press in mathematical physics.
On an inner product space, bounded linear operators admit adjoints: for T a bounded operator, the adjoint T* satisfies ⟨Tx,y⟩ = ⟨x,T*y⟩. Self-adjoint (Hermitian) operators T = T* model physical observables in quantum theory, while unitary operators model time evolution and symmetries (Stone's theorem connects one-parameter unitary groups to self-adjoint generators). Concepts of orthogonal (or unitary) projections, orthonormal bases, and the Gram–Schmidt process rely on the inner product. Domain issues for unbounded operators (e.g., momentum and position operators) are treated using densely defined operators in Hilbert spaces and techniques from functional analysis.
Quantum mechanics represents pure states by rays in a Hilbert space; the inner product between normalized vectors gives complex probability amplitudes. The Dirac bra–ket notation explicitly uses the inner product: ⟨φ|ψ⟩ denotes the inner product of |φ⟩ and |ψ⟩. Born's rule states that measurement probabilities are |⟨φ|ψ⟩|^2. Observables correspond to self-adjoint operators on the Hilbert space, and expectation values are given by ⟨ψ|A|ψ⟩. Prominent formulations by Paul Dirac, John von Neumann, and later authors formalized these links between inner products and measurement statistics used in laboratories like CERN and quantum information groups at IBM and Google.
The spectral theorem for self-adjoint (and normal) operators on a Hilbert space provides a decomposition into projection-valued measures and underlies quantum measurement theory: observables have spectral measures that determine possible outcomes and their probabilities. Continuous spectra (e.g., position operator) require generalized eigenvectors and rigged Hilbert space techniques developed by Maurice Gel'fand and collaborators. The spectral calculus, functional calculus, and projection-valued measures are central to rigorous derivations of time evolution, scattering theory, and the formalism of quantum statistical mechanics in works by E. H. Lieb and O. E. Lanford.
The inner product structure extends to tensor products: given two inner product spaces V and W, the tensor product V ⊗ W carries a natural inner product making simple tensors orthonormal when factors are orthonormal. This construction models composite quantum systems, entanglement, and joint observables; separable Hilbert spaces lead to standard constructions in quantum information theory and the study of Bell inequalities, deployed in experiments by groups at University of Innsbruck and MIT. Properties of partial trace, Schmidt decomposition (a consequence of singular value decomposition), and entanglement measures rely on the inner product and associated operator theory.