10.3 Normal Operators and Unitary Operators

Normal and unitary operators are key players in the world of linear algebra. They're like the cool kids of operators, with special properties that make them super useful in quantum mechanics and other fields.

These operators build on what we've learned about adjoints and self-adjoint operators. They have unique traits that set them apart, like commuting with their adjoints or preserving inner products. Understanding them is crucial for grasping advanced linear algebra concepts.

Normal Operators in Inner Product Spaces

Definition and Properties

• A normal operator is a bounded linear operator $T$ on a Hilbert space $H$ such that $T$ commutes with its adjoint $T^*$, satisfying the equation $TT^* = T^*T$
• Normal operators have a complete set of orthonormal eigenvectors that span the Hilbert space
  • This means that any vector in the Hilbert space can be expressed as a linear combination of the eigenvectors of the normal operator
• The eigenvalues of a normal operator are complex numbers, and the eigenvectors corresponding to distinct eigenvalues are orthogonal
  • If $\lambda_1$ and $\lambda_2$ are distinct eigenvalues of a normal operator, then their corresponding eigenvectors $v_1$ and $v_2$ satisfy $\langle v_1, v_2 \rangle = 0$
• The norm of a normal operator is equal to its spectral radius, which is the maximum absolute value of its eigenvalues
  • For a normal operator $T$, $\|T\| = \max\{|\lambda| : \lambda \text{ is an eigenvalue of } T\}$

Spectral Theorem Characterization

• Normal operators are characterized by the spectral theorem, which states that they can be represented as a linear combination of orthogonal projections onto the eigenspaces corresponding to their eigenvalues
  • If $T$ is a normal operator with eigenvalues $\lambda_1, \lambda_2, \ldots, \lambda_n$ and corresponding eigenspaces $E_1, E_2, \ldots, E_n$, then $T = \sum_{i=1}^n \lambda_i P_i$, where $P_i$ is the orthogonal projection onto $E_i$
• The spectral theorem provides a powerful tool for analyzing and understanding the behavior of normal operators in inner product spaces
  • It allows for the decomposition of a normal operator into a sum of simpler operators, each associated with a specific eigenvalue and eigenspace
• The spectral theorem also implies that the eigenspaces of a normal operator are mutually orthogonal and span the entire Hilbert space
  • This orthogonality property is crucial for many applications, such as quantum mechanics, where the eigenstates of an observable (represented by a normal operator) are required to be orthogonal

Normal vs Self-Adjoint vs Unitary Operators

Self-Adjoint Operators (Hermitian Operators)

• Self-adjoint operators (also known as Hermitian operators) are a special case of normal operators, where $T = T^*$
  • For a self-adjoint operator, the adjoint is equal to the original operator
• The eigenvalues of self-adjoint operators are always real, and their eigenvectors form an orthonormal basis for the Hilbert space
  • This property makes self-adjoint operators particularly useful in quantum mechanics, where observables are represented by self-adjoint operators, and their eigenvalues correspond to the possible measurement outcomes
• Examples of self-adjoint operators include real symmetric matrices and the position and momentum operators in quantum mechanics

Unitary Operators

• Unitary operators are another special case of normal operators, where $T^* = T^{-1}$
  • The adjoint of a unitary operator is equal to its inverse
• Unitary operators have complex eigenvalues with absolute value 1, and their eigenvectors form an orthonormal basis for the Hilbert space
  • The eigenvalues of a unitary operator lie on the unit circle in the complex plane
• Unitary operators preserve the inner product and norm of vectors, making them important in quantum mechanics for describing the evolution of quantum states
  • Examples of unitary operators include rotation matrices and the time-evolution operator in quantum mechanics

Relationship between Normal, Self-Adjoint, and Unitary Operators

• Every normal operator can be decomposed into the sum of a self-adjoint operator and an imaginary multiple of a self-adjoint operator
  • If $T$ is a normal operator, then $T = A + iB$, where $A$ and $B$ are self-adjoint operators
  • This decomposition is known as the Cartesian decomposition of a normal operator
• The product of two normal operators is normal if and only if they commute
  • If $S$ and $T$ are normal operators, then $ST$ is normal if and only if $ST = TS$
  • This property is important for understanding the behavior of composite quantum systems and the construction of tensor product spaces

Spectral Theorem for Normal Operators

Spectral Decomposition

• The spectral theorem states that for any normal operator $T$ on a Hilbert space $H$, there exists a unique resolution of the identity $E$ (a family of orthogonal projections) such that $T$ can be represented as the integral of the identity with respect to $E$
  • Mathematically, this can be expressed as $T = \int_{\sigma(T)} \lambda dE(\lambda)$, where $\sigma(T)$ is the spectrum of $T$ (the set of all eigenvalues)
• The spectral theorem implies that any normal operator has a spectral decomposition, i.e., it can be written as a linear combination of orthogonal projections onto its eigenspaces
  • If $T$ is a normal operator with eigenvalues $\lambda_1, \lambda_2, \ldots, \lambda_n$ and corresponding eigenspaces $E_1, E_2, \ldots, E_n$, then $T = \sum_{i=1}^n \lambda_i P_i$, where $P_i$ is the orthogonal projection onto $E_i$

Eigenvalues and Eigenvectors

• The eigenvalues of a normal operator are the values $\lambda$ for which the operator $T - \lambda I$ is not invertible, where $I$ is the identity operator
  • In other words, $\lambda$ is an eigenvalue of $T$ if and only if there exists a non-zero vector $v$ such that $Tv = \lambda v$
• The eigenvectors of a normal operator corresponding to distinct eigenvalues are orthogonal, and the eigenvectors corresponding to the same eigenvalue form an orthonormal basis for the corresponding eigenspace
  • If $v_1$ and $v_2$ are eigenvectors of $T$ corresponding to distinct eigenvalues $\lambda_1$ and $\lambda_2$, then $\langle v_1, v_2 \rangle = 0$
  • If $v_1, v_2, \ldots, v_k$ are eigenvectors of $T$ corresponding to the same eigenvalue $\lambda$, then they can be chosen to form an orthonormal basis for the eigenspace associated with $\lambda$

Applications of Unitary Operators

Preservation of Inner Product and Norm

• Unitary operators preserve the inner product between vectors, i.e., $\langle Ux, Uy \rangle = \langle x, y \rangle$ for all $x, y$ in the Hilbert space $H$
  • This property is essential for maintaining the geometry of the Hilbert space and the relationships between vectors under unitary transformations
• Unitary operators are isometries, meaning they preserve the norm of vectors: $\|Ux\| = \|x\|$ for all $x$ in $H$
  • This property ensures that unitary operators do not stretch or shrink vectors, but only rotate or reflect them in the Hilbert space

Composition and Inversion

• The composition of two unitary operators is also a unitary operator, and the inverse of a unitary operator is its adjoint
  • If $U$ and $V$ are unitary operators, then $UV$ is also a unitary operator
  • The inverse of a unitary operator $U$ is given by its adjoint $U^*$, satisfying $UU^* = U^*U = I$
• These properties allow for the construction of complex unitary transformations by combining simpler unitary operators
  • For example, the Hadamard gate in quantum computing can be expressed as the product of simpler unitary matrices

Change of Basis

• Unitary operators can be used to define a change of basis in a Hilbert space, as they map one orthonormal basis to another
  • If $\{e_1, e_2, \ldots, e_n\}$ is an orthonormal basis for a Hilbert space $H$ and $U$ is a unitary operator, then $\{Ue_1, Ue_2, \ldots, Ue_n\}$ is also an orthonormal basis for $H$
• Change of basis is particularly useful in quantum mechanics, where different bases correspond to different measurement scenarios or different representations of a quantum system
  • For example, the Fourier basis is often used in quantum algorithms to perform operations in the frequency domain

Matrix Representation

• The matrix representation of a unitary operator with respect to an orthonormal basis is a unitary matrix, i.e., a complex square matrix $U$ satisfying $U^*U = UU^* = I$
  • If $U$ is a unitary operator and $\{e_1, e_2, \ldots, e_n\}$ is an orthonormal basis for a Hilbert space $H$, then the matrix elements of $U$ with respect to this basis are given by $u_{ij} = \langle e_i, Ue_j \rangle$
• Unitary matrices have important applications in various fields, such as quantum computing, where they represent quantum gates, and in signal processing, where they are used for transformations like the discrete Fourier transform
  • The Pauli matrices and the Hadamard matrix are examples of commonly used unitary matrices in quantum computing