Alphabeta Math
TheoremStatement: AI-adaptedProof: AI-adaptedSession-authored (Fable 5 assisted)precheck passjudge pass (deepseek-v4-pro + gpt-5.6-terra)audited 2026-08-16
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.

Power ranks determine every nilpotent Jordan-block multiplicity

Statement

Let N be nilpotent on a finite-dimensional vector space. Put dk=dimkerNk and ρk=rankNk for k0, so d0=0 and ρ0=dimV. For every k1, #{blocks of size at least k}=dkdk1=ρk1ρk, and #{blocks of size exactly k}=2dkdk1dk+1=ρk12ρk+ρk+1. Thus either the nullities or the ranks of all powers determine the multiset of nilpotent Jordan blocks. On the zero space all sequences are zero and the block multiset is empty.

Facts & Assumptions

Given: A nilpotent endomorphism N of a finite-dimensional vector space.

[L1]

There is a basis in which N is a direct sum of nilpotent Jordan blocks (Every finite-dimensional nilpotent endomorphism has a basis of Jordan strings).

[L2]

For every k0, kerTkkerTk+1 and imTk+1imTk; and if kerTm=kerTm+1 for some m0 — equivalently rankTm=rankTm+1 — then kerTm+r=kerTm and imTm+r=imTm for every r0 (Kernel and rank sequences of powers stabilise once equality occurs).

[L3]

Nullity and rank are the dimensions of the kernel and image (Rank and nullity of a linear map with finite-dimensional domain).

[L4]

Rank-nullity gives dimV=dimkerS+dimimS for every endomorphism S of V (Rank-nullity: dimFV=nullityT+rankT).

Proof

technique · direct
1.1

On a block Jm(0), dimkerJm(0)k=min(k,m); its contribution to dkdk1 is therefore 1 exactly when mk, and 0 otherwise. Summing across the block basis from [L1] gives the first nullity formula.

L1L3algebra
2.1

Subtracting the number of blocks of size at least k+1 from the number of blocks of size at least k gives 2dkdk1dk+1 blocks of exact size k.

step 1.1algebra
3.1

Fact [L4] gives dk+ρk=dimV for every k, so replacing each dk in steps 1.1-2.1 gives the two rank formulas. For the tail, first dispose of V=0: there the block multiset is empty, every dk and ρk is zero, and both formulas read 0=0. So assume V0, in which case the basis of [L1] has at least one block and a largest block size m exists; step 1.1 gives dimkerJm(0)k=min(k,m)=m for every block and every km, so dk=dimV and ρk=0 for all km. In particular kerNm=kerNm+1, which is the hypothesis of [L2], and [L2] then gives the stabilised tail beyond the largest block.

step 1.1step 2.1L1L2L3L4algebra
4.1

These formulas recover every block multiplicity, including size one and the endpoint after the largest block; when V=0 each quantity and each recovered multiplicity is zero.

step 1.1step 2.1step 3.1

Depends on

Used by

Dependency tree · next 3 levels

Direct dependencies and their dependencies through the next three levels: 54 results over 16 levels. An arrow runs from a result to what uses it, and this result sits at the bottom with a heavier outline. Click the chart to enlarge it.

Sources