Alphabeta Math
AlgorithmAlgorithm: Literature-sourcedProof: Not applicablePipeline-generatedjudge 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.

Exact finite Francis single- or double-shift Hessenberg QR step

Definition

Input an exact upper-Hessenberg matrix H∈Fn×n, with F=C or R and n≥1, and exactly one shift descriptor:

  1. Complex single shift: F=C and μ∈C. Set p(z)=z−μ and r=1.
  2. Real double shift: F=R, n≥2, and p(z)=z2−sz+t with s,t∈R. In the ordinary Francis branch, s and t are the trace and determinant of the trailing 2×2 block; a declared exceptional shift may supply another real pair. Set r=2.

Form x=p(H)e1. For a Hessenberg H, x is supported in its first r+1 entries; compute only those entries. If x=0, return H+=H, Q=I, and status zero-shift-vector. Otherwise choose a Householder reflector P0 on the first min⁡(r+1,n) coordinates with P0x=αe1, embed it in Fn×n, and set H1=P0∗HP0, Q=P0.

For j=1,…,max⁡(n−2,0), in increasing order, let Ij={j+1,…,min⁡(j+r+1,n)}. Thus Ij has at most three indices in a real double shift and at most two in a complex single shift. Extract y=(Hj)Ij,j. If y=0, use the identity on Ij; otherwise choose a Householder reflector Pj on Ij sending y to a multiple of its first coordinate vector. Embed Pj in the identity, update Hj+1=Pj∗HjPj, and append it to Q on the right. After the finite chase, check the exact matrix zero pattern and the first-column condition. Return completed only if the output is upper Hessenberg and Qe1 is proportional to p(H)e1; otherwise return bulge-chase-invariant-failure with the partial similarity factors. The procedure makes no convergence claim.

Depends on

Used by

Dependency tree · two levels

11 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