Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-adaptedPipeline-generatedprecheck passjudge pass (gpt-6-luna)
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Finite Hermitian ground vectors and a deterministic tie rule

Statement

Let F∈{R,C}, let m≥1, and let A∈Fm×m satisfy A∗=A. Define qA(x):=x∗Axx∗x(x∈Fm∖{0}). Then qA has a minimum λmin⁡, equal to the least eigenvalue of A, and its minimizers are precisely the nonzero vectors in the lowest eigenspace Emin⁡=ker⁡(A−λmin⁡I). There is a deterministic choice of a unit minimizer: let Pmin⁡ be the orthogonal projection onto Emin⁡, scan the standard basis e1,…,em, and take the first j with Pmin⁡ej≠0. Normalize Pmin⁡ej; over C, if z is its first nonzero coordinate, multiply by z‾/∣z∣, and over R multiply by z/∣z∣.

Facts & Assumptions

Given: A real symmetric or complex Hermitian matrix A∈Fm×m with m≥1.

[F1]

Self-adjoint and normal endomorphisms of a finite-dimensional real or complex inner product space: self-adjoint means T∗=T, and every self-adjoint endomorphism is normal.

[F2]

In an orthonormal basis, self-adjoint means conjugate-transpose symmetry and normal means commuting with the conjugate transpose: in an orthonormal basis, self-adjointness is equivalent to the matrix identity A=A∗.

[F3]

The Rayleigh quotient of a nonzero vector for a self-adjoint endomorphism: on a finite-dimensional real inner-product space, RT(x)=⟨Tx,x⟩/⟨x,x⟩ for x≠0.

[F4]

Courant-Fischer min-max principle for self-adjoint endomorphisms on finite-dimensional real inner product spaces: for a self-adjoint endomorphism of an m-dimensional real inner-product space, the least eigenvalue is the minimum of its Rayleigh quotient over nonzero vectors.

[F5]
[F6]

Complex spectral theorem: a normal endomorphism of a finite-dimensional complex inner product space has an orthonormal eigenbasis, and conversely: a normal endomorphism of a finite-dimensional complex inner-product space has an orthonormal eigenbasis.

[F7]

For a subspace W of a finite-dimensional inner product space, V=W⊕W⊥: a subspace of a finite-dimensional inner-product space has an orthogonal complement, and every vector has a unique decomposition into the subspace and its orthogonal complement.

Proof

1.1F2F3F4F5algebra

Suppose F=R. By [F2], A represents a self-adjoint endomorphism in the standard orthonormal basis, and [F3] identifies its Rayleigh quotient with qA. Taking k=m in [F4] gives λmin⁡=min⁡dim⁡W=1max⁡x∈W∖{0}qA(x). On each one-dimensional W, homogeneity gives the same quotient for every nonzero vector, and every nonzero x spans such a W; hence this minimum is exactly min⁡x≠0qA(x). By [F5], write x=∑jcjuj in an orthonormal eigenbasis with real eigenvalues λj; then qA(x)=∑jλjcj2/∑jcj2, which equals λmin⁡ exactly when every nonzero coefficient belongs to the least-eigenvalue eigenspace.

1.2F1F2F6algebra

Suppose F=C. By [F2], A represents a self-adjoint endomorphism in the standard orthonormal basis; by [F1] this endomorphism is normal, so [F6] gives an orthonormal eigenbasis u1,…,um. Every eigenvalue is real because if Auj=λjuj, then uj∗Auj‾=uj∗A∗uj=uj∗Auj, so λj∥uj∥2 is real and ∥uj∥2>0. For x=∑jcjuj≠0, the quotient is the weighted average ∑jλj∣cj∣2/∑j∣cj∣2, so it is at least the least eigenvalue and equals it exactly when all nonzero coefficients lie in its eigenspace.

2.1F2F4F5F6F7step 1.1step 1.2algebra∎

In either field, the least eigenspace is nonzero, so [F7] gives a unique orthogonal decomposition of each vector into its Emin⁡ and Emin⁡⊥ parts; define Pmin⁡ to be the first component. Uniqueness makes Pmin⁡ linear, since sums and scalar multiples of decompositions are the corresponding decompositions. At least one Pmin⁡ej is nonzero, since otherwise linearity and the fact that the standard vectors span Fm would imply Pmin⁡=0, contradicting Emin⁡≠{0}. The first such j is uniquely determined by the fixed basis order. Normalize this projection. Over C, its first nonzero coordinate z is changed to ∣z∣>0 by multiplication by z‾/∣z∣; over R, multiplication by z/∣z∣ makes that coordinate positive. These unit-modulus rescalings preserve the eigenvector and its Rayleigh quotient. The minimizer characterization in steps 1.1 and 1.2 proves the result.

Depends on

Used by

Dependency tree · two levels

19 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources