Skip to article frontmatterSkip to article content
Tutorial

The Adjoint

Abstract

The definition of the adjoint operator.

Keywords:VectorsVector SpacesInner ProductsAdjointTransposeComplex Conjugate.

Adjoint Definition

Let VV be an inner product space defined over C\mathbb{C}, with inner product (,):CnCnC1(\cdot, \cdot): \mathbb{C}^{n} \otimes \mathbb{C}^{n} \rightarrow \mathbb{C}^{1}, and L:VVL: V \rightarrow V a linear operator.

The adjoint operator of the linear operator LL, denoted by LL^{\dagger}, is the operator that satisfies the following relation

(L(x),y)=(x,L(y)),(L^{\dagger}(\bm{x}), \bm{y}) = (\bm{x}, L(\bm{y})),

where x,y  V\bm{x}, \bm{y}~\in~V.

Associated Matrix of the Adjoint

Let MLBM^{\mathfrak{B}}_{L} be the matrix associated to the linear map LL with respect to the orthonormal basis B\mathfrak{B} and the element in the iith row and jjth colunm be aija_{ij}.

The matrix associated to the linear map LL^{\dagger} with respect to the orthonormal basis B\mathfrak{B}, MLBM_{L^{\dagger}}^{\mathfrak{B}}, is given by taking the complex conjugate of all elements of MLBM_{L}^{\mathfrak{B}} and then transposing. Hence, the element in the iith row and jjth column of MLBM_{L^{\dagger}}^{\mathfrak{B}} will be ajia^{*}_{ji}.

For this reason, the adjoint is also sometimes referred to as the conjugate transpose operations.

Properties of the Adjoint

Let V,(,)V, (\cdot, \cdot) be an inner product space defined over C\mathbb{C}, let L,T:VVL, T: V \rightarrow V be linear operators and λ  C1\lambda~\in~\mathbb{C}^{1}.

Matrix Properties From the Adjoint

Unitary
Hermitian
Normal

Let A  MnmA~\in~\mathbb{M}_{nm}.

AA is unitary if AA=AA=IA^{\dagger}A=AA^{\dagger}=\mathbb{I},

where I\mathbb{I} is the identity.

Properties

Let AA be unitary, then:

  • A1A^{-1} is unitary.
  • Ux=x\vert \vert U \bm{x} \vert \vert = \vert \vert \bm{x} \vert \vert, where x  V\bm{x}~\in~V and \vert \vert \cdot \vert \vert is a norm.
  • If λ \lambda is an eigenvalue of AA, then λ=1\vert \lambda \vert = 1, meaning AA has complex eigenvalues of magnitude one.
  • The columns of AA form an orthonormal basis.