• Definition (Linear Map): A map T : V → W is linear if T (u + v) = T (u) + T (v), T (λv) = λT (v) for all u, v ∈ V and scalars λ. Key point: Verify linearity by checking these two properties. • Injective ⇐⇒ Null(T ) = {0}: To show T is injective, show T (v) = 0 =⇒ v = 0. • Invertibility: For T : V → W finite-dimensional with dim V = dim W , T invertible ⇐⇒ T injective ⇐⇒ T surjective. Key point: Proving injectivity often suffices for invertibility. • Isomorphism & Dimension: Two finite-dimensional vector spaces are isomorphic if and only if they have the same dimension. • Given bases v1 , . . . , vn of V and w1 , . . . , wm of W , each T ∈ L(V, W ) corresponds to a matrix [T ] ∈ Fm×n . The map T 7→ [T ] is an isomorphism L(V, W ) ∼ = Fm×n . • Over C, every linear operator T : V → V is unitarily upper-triangularizable (Schur’s Theorem). • Eigenvalues: λ is an eigenvalue of T if (T − λI) is not invertible. Equivalently, ∃v ̸= 0 : T (v) = λv. • Existence (Complex Case): Every operator on a nonzero complex vector space has at least one eigenvalue. • Diagonalizability: T is diagonalizable ⇐⇒ V can be expressed as a direct sum of its eigenspaces. More explicitly, if T is diagonalizable and has eigenvalues λ1 , . . . , λk , then V = E(λ1 , T ) (direct sum) E(λ2 , T ) (direct sum) · · · (direct sum) E(λk , T ). If T has dim V distinct eigenvalues, it is diagonalizable. (Note: “Direct sum” means every element in V can be uniquely written as a sum of elements from each of these subspaces, and the only vector common to all is the zero vector.) • Minimal Polynomial: For a polynomial q, q(T ) = 0 iff q is a multiple of the minimal polynomial of T . The minimal polynomial governs factors and nilpotent parts. • Nilpotent Operators: T nilpotent ⇐⇒ the only eigenvalue is 0 ⇐⇒ minimal polynomial is z m . Such an operator can be represented by a strictly upper-triangular matrix in some basis. 1 • Generalized Eigenvectors & Jordan Decomposition: Over C, V = G(λ1 , T ) (direct sum) G(λ2 , T ) · · · (direct sum) G(λm , T ), where each G(λi , T ) is a generalized eigenspace. • Characteristic Polynomial: Over C, pT (z) = (z − λ1 )d1 · · · (z − λm )dm . Cayley-Hamilton Theorem: pT (T ) = 0. • Inner Product: ⟨·, ·⟩ is linear in the first argument, conjugate symmetric, and positive definite. Use it to define norms and check orthogonality. • Orthonormality: Orthonormal sets are linearly independent. Every finite-dimensional inner product space has an orthonormal basis (Gram-Schmidt). • Orthogonal Projections: For a subspace U , V = U (direct sum) U ⊥ . The orthogonal projection PU satisfies PU2 = PU , range(PU ) = U , and gives the best approximation in U . • Adjoint (T ∗ ): Defined by ⟨T v, w⟩ = ⟨v, T ∗ w⟩. Properties: (ST )∗ = T ∗ S ∗ and (T ∗ )∗ = T . • Null/Range Relations: Null(T ∗ ) = (range(T ))⊥ , Range(T ∗ ) = (Null(T ))⊥ . • Self-Adjoint: T = T ∗ ; all eigenvalues are real; T is orthogonally diagonalizable. • Normal: T T ∗ = T ∗ T . Normal operators are diagonalizable with an orthonormal basis. If T is normal, eigenvectors from distinct eigenvalues are orthogonal. • Spectral Theorems: Over R : T self-adjoint ⇐⇒ T is orthogonally diagonalizable. Over C : T normal ⇐⇒ T is unitarily diagonalizable. Tips • Dimension Arguments: If dim V = dim W , showing injectivity or surjectivity alone implies invertibility. • Minimal/Characteristic Polynomials: Use them to deduce eigenvalues, nilpotency, and diagonalizability quickly. • Orthonormal Bases: Simplify computations and proofs involving inner products and adjoints. • Normal & Self-Adjoint Operators: Diagonalizability with an orthonormal basis is guaranteed. For self-adjoint operators, the spectrum is real. 2