How does a Hilbert space differ from a Banach space?

A Hilbert space carries an inner product that induces its norm and supplies geometry, angles, orthogonality, and projection, whereas a general Banach space has only a norm; every Hilbert space is a Banach space but not conversely.

What is an orthonormal basis?

It is a maximal set of mutually perpendicular unit vectors such that every element of the space is the sum of its projections onto them, generalizing the way Fourier series expand functions in sines and cosines.

Hilbert Spaces

A Hilbert space is a complete inner-product space, an infinite-dimensional generalization of Euclidean geometry where notions of angle, orthogonality, and projection retain their full power.

Definition

A Hilbert space is a vector space with an inner product that is complete in the norm the inner product induces; the inner product supplies a geometry of lengths and angles that makes orthogonal projection and orthonormal expansion available.

Scope

This topic covers the inner product and its induced norm, the Cauchy-Schwarz and parallelogram identities, orthogonality and orthogonal complements, the projection theorem onto closed convex sets, orthonormal bases and Parseval's identity, and the Riesz representation theorem identifying a Hilbert space with its dual.

Core questions

How does an inner product equip an infinite-dimensional space with geometry?
Why does every closed convex set admit a unique nearest point, and what does this projection give?
How do orthonormal bases represent every vector as a generalized Fourier series?
Why is a Hilbert space naturally identified with its own dual?

Key theories

Projection theorem: Every nonempty closed convex subset of a Hilbert space contains a unique point nearest to any given vector, and orthogonal projection onto a closed subspace splits the space into the subspace and its orthogonal complement.
Riesz representation theorem: Every bounded linear functional on a Hilbert space is given by the inner product with a unique vector, so the space is isometrically identified with its dual, the source of much of the space's analytic convenience.

Clinical relevance

Hilbert spaces are the state spaces of quantum mechanics, where orthonormal expansion and projection express measurement and superposition; they also underlie least-squares approximation, Fourier and wavelet analysis, signal processing, and the reproducing-kernel spaces central to modern machine learning.

History

The structure emerged from Hilbert's study of integral equations and infinite quadratic forms in the early twentieth century; von Neumann gave the abstract axiomatic definition in the 1920s while formulating quantum mechanics, fixing the modern notion of a Hilbert space.

Key figures

David Hilbert
John von Neumann
Frigyes Riesz

Seminal works

conway1985
stein2005real

Frequently asked questions

How does a Hilbert space differ from a Banach space?: A Hilbert space carries an inner product that induces its norm and supplies geometry, angles, orthogonality, and projection, whereas a general Banach space has only a norm; every Hilbert space is a Banach space but not conversely.
What is an orthonormal basis?: It is a maximal set of mutually perpendicular unit vectors such that every element of the space is the sum of its projections onto them, generalizing the way Fourier series expand functions in sines and cosines.