Eigenvectors for distinct eigenvalues are linearly independent
ProvedFamousTheorems.eigenvectors_linear_independent_7bEigenvectors for distinct eigenvalues are linearly independent. Let be a linear endomorphism of a vector space over a field . Let be a set of scalars and, for each , let be an eigenvector of with eigenvalue . Then the family is linearly independent.
It follows that an endomorphism of an -dimensional space has at most eigenvalues, and that it is diagonalizable if it has distinct eigenvalues. The standard proof takes a shortest linear relation and applies to reduce its length.
Formalization note. Mathlib's Module.End.eigenvectors_linearIndependent, specialized from integral domains to fields. f.HasEigenvector μ x says that and . The family is indexed by the elements of viewed as a type.
Preamble
import Mathlib
Formal statement
namespace FamousTheorems
theorem eigenvectors_linear_independent_7b {K V : Type*} [Field K] [AddCommGroup V] [Module K V] (f : Module.End K V) (μs : Set K)
(xs : μs → V) (h : ∀ μ : μs, f.HasEigenvector μ (xs μ)) : LinearIndependent K xs := by sorry
end FamousTheoremsSource
Listed in Mathlib's curated theorem manifests (docs/1000.yaml); formalized in Mathlib as `Module.End.eigenvectors_linearIndependent`. Proof here reduces to that Mathlib result.