Functional Analysis

🧐Functional Analysis Unit 2 – Linear Operators in Normed Spaces

Linear operators in normed spaces are fundamental to functional analysis. They map elements between vector spaces while preserving linear combinations, with properties like linearity, injectivity, and surjectivity shaping their behavior. Bounded linear operators have finite amplification factors and form vector spaces. Operator norms quantify their size, while continuity and boundedness are equivalent in normed spaces. These concepts underpin advanced topics like spectral theory and operator algebras.

Key Concepts and Definitions

  • Linear operators map elements from one vector space to another while preserving linear combinations
  • Normed spaces consist of a vector space equipped with a norm that measures the "size" of vectors
  • Domain and codomain refer to the input and output spaces of a linear operator, respectively
  • Null space (kernel) contains all vectors mapped to the zero vector by the operator
    • Dimension of the null space is called the nullity of the operator
  • Range (image) is the set of all vectors in the codomain that are outputs of the operator
    • Dimension of the range is called the rank of the operator

Properties of Linear Operators

  • Linearity ensures that the operator preserves vector addition and scalar multiplication
    • T(u+v)=T(u)+T(v)T(u + v) = T(u) + T(v) for all vectors uu and vv in the domain
    • T(αu)=αT(u)T(αu) = αT(u) for all scalars αα and vectors uu in the domain
  • Injectivity (one-to-one) means distinct inputs map to distinct outputs
  • Surjectivity (onto) means every vector in the codomain is an output of the operator
  • Bijectivity combines injectivity and surjectivity, implying a one-to-one correspondence between domain and codomain
  • Invertibility allows for the existence of an inverse operator that "undoes" the original operator
    • Invertible operators are always bijective

Bounded Linear Operators

  • Bounded linear operators have a finite "amplification factor" for all input vectors
    • There exists a constant M0M \geq 0 such that T(u)Mu\|T(u)\| \leq M\|u\| for all uu in the domain
  • Boundedness is a stronger condition than continuity for linear operators
    • All bounded linear operators are continuous, but not all continuous linear operators are bounded
  • Bounded linear operators form a vector space, allowing for addition and scalar multiplication of operators
  • Composition of bounded linear operators is also a bounded linear operator
  • Inverse of a bounded linear operator (if it exists) is also bounded

Operator Norms

  • Operator norms quantify the "size" or "amplification factor" of a linear operator
    • Defined as T=sup{T(u):u=1}\|T\| = \sup\{\|T(u)\| : \|u\| = 1\}, where uu is in the domain
  • Operator norms satisfy the properties of a norm on the space of bounded linear operators
    • Non-negativity: T0\|T\| \geq 0, with equality if and only if T=0T = 0
    • Homogeneity: αT=αT\|αT\| = |α|\|T\| for all scalars αα
    • Triangle inequality: T+ST+S\|T + S\| \leq \|T\| + \|S\| for all bounded linear operators TT and SS
  • Commonly used operator norms include the operator norm induced by the p\ell^p norms on the domain and codomain spaces

Examples and Applications

  • Differentiation operator maps functions to their derivatives in function spaces (Sobolev spaces)
  • Integration operator maps functions to their integrals (Lebesgue spaces)
  • Fourier transform is a linear operator that maps functions to their frequency representations
    • Plays a crucial role in signal processing and quantum mechanics
  • Rotation matrices are linear operators that rotate vectors in Euclidean spaces
  • Projection operators map vectors onto subspaces (orthogonal projections in Hilbert spaces)
  • Compact operators generalize the notion of matrices to infinite-dimensional spaces
    • Integral operators and Hilbert-Schmidt operators are examples of compact operators

Continuity and Boundedness

  • Continuity of linear operators is equivalent to boundedness in normed spaces
    • A linear operator TT is continuous if and only if it is bounded
  • Continuous linear operators map convergent sequences to convergent sequences
    • If unuu_n \to u in the domain, then T(un)T(u)T(u_n) \to T(u) in the codomain
  • Uniform boundedness principle states that a family of pointwise bounded linear operators is uniformly bounded
  • Open mapping theorem asserts that a surjective bounded linear operator maps open sets to open sets
  • Closed graph theorem characterizes continuous linear operators in terms of their graphs

Operator Algebras

  • Operator algebras study the algebraic and topological properties of collections of linear operators
  • Banach algebras are normed algebras that are complete as metric spaces
    • Space of bounded linear operators on a Banach space forms a Banach algebra
  • C*-algebras are Banach algebras equipped with an involution (adjoint operation) satisfying certain axioms
    • Provide a framework for studying quantum mechanics and non-commutative geometry
  • Von Neumann algebras (W*-algebras) are C*-algebras that are closed in the weak operator topology
    • Play a central role in the mathematical formulation of quantum mechanics

Advanced Topics and Extensions

  • Spectral theory studies the eigenvalues and eigenvectors of linear operators
    • Spectrum of an operator generalizes the notion of eigenvalues to infinite-dimensional spaces
  • Functional calculus allows for the definition of functions of operators
    • Holomorphic functional calculus extends this notion to analytic functions
  • Fredholm theory deals with the solvability of linear equations involving compact perturbations of the identity
    • Fredholm alternative characterizes the existence and uniqueness of solutions
  • Toeplitz operators are a class of operators that arise in the study of analytic functions (Hardy spaces)
  • Pseudodifferential operators generalize differential operators and have applications in partial differential equations
  • Operator theory finds applications in quantum mechanics, signal processing, and partial differential equations


© 2024 Fiveable Inc. All rights reserved.
AP® and SAT® are trademarks registered by the College Board, which is not affiliated with, and does not endorse this website.

© 2024 Fiveable Inc. All rights reserved.
AP® and SAT® are trademarks registered by the College Board, which is not affiliated with, and does not endorse this website.