This chapter investigates the effects of adding the additional structure of an inner product to a finite-dimensional real or complex vector space.
Section 1 concerns the effect on the vector space itself, defining inner products and their corresponding norms and giving a number of examples and formulas for the computation of norms. Vector-space bases that are orthonormal play a special role.
Section 2 concerns the effect on linear maps. The inner product makes itself felt partly through the notion of the adjoint of a linear map. The section pays special attention to linear maps that are self-adjoint, i.e., are equal to their own adjoints, and to those that are unitary, i.e., preserve norms of vectors.
Section 3 proves the Spectral Theorem for self-adjoint linear maps on finite-dimensional innerproduct spaces. The theorem says in part that any self-adjoint linear map has an orthonormal basis of eigenvectors. The Spectral Theorem has several important consequences, one of which is the existence of a unique positive semidefinite square root for any positive semidefinite linear map. The section concludes with the polar decomposition, showing that any linear map factors as the product of a unitary linear map and a positive semidefinite one.
KeywordsOrthonormal Basis Hermitian Matrix Vector Subspace Polar Decomposition Complex Vector Space
Unable to display preview. Download preview PDF.