Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-generatedprecheck passjudge pass (z-ai/glm-5.2)audited 2026-07-28
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 sums re-indexed along an injection, with a zero term deleted, and concatenated; and the closure properties of linear independence: an independent list is injective and never 0V, its sublists are independent, a list is independent exactly when it is injective with linearly independent image, and every subset of a linearly independent set is linearly independent

Statement

Let V be a vector space over a field F (Vector space over a field), with finite sums of vectors as in Linear combination of a finite list, and the span span⁡(S) as the smallest linear subspace containing S and linear independence as in Linear independence: a finite list v:n→V is independent when ∑i<nλivi=0V forces every λi=0F, and a subset S⊆V is independent when every injective finite list into S is independent. For a function f and a set A we write f[A] for the image of A (Injection, surjection, bijection).

Three facts about finite sums.

  1. Re-indexing along an injection. Let n,m∈N, let ι:m→n be injective, and let u:n→V satisfy uj=0V for every j<n with j∉ι[m]. Then ∑j<nuj  =  ∑i<muι(i).
  2. Deleting one index. Let n′∈N and k<σ(n′). The map δk:n′→σ(n′) given by δk(i)=i for i<k and δk(i)=σ(i) for k≤i<n′ is injective with image σ(n′)∖{k}. Consequently, if u:σ(n′)→V has uk=0V, then ∑j<σ(n′)uj=∑i<n′uδk(i).
  3. Concatenation. Let a,q∈N, y:a→V and z:q→V. There is exactly one list c:a+q→V with ci=yi for i<a and ca+j=zj for j<q, and it satisfies ∑i<a+qci  =  ∑i<ayi  +  ∑j<qzj. If moreover y and z are injective with y[a]∩z[q]=∅, then c is injective with image y[a]∪z[q].

Four facts about independence.

  1. Every linearly independent list v:n→V is injective, and vi≠0V for every i<n.
  2. If v:n→V is linearly independent and ι:m→n is injective, then the sublist v∘ι:m→V is linearly independent.
  3. A list v:n→V is linearly independent if and only if it is injective and its image v[n] is a linearly independent subset of V; in that case v is a bijection n→v[n], so v[n]≈n (Equinumerous sets, A≈B and A⪯B).
  4. Every subset of a linearly independent subset of V is linearly independent.

Facts & Assumptions

Given: A field F, a vector space V over F, and the finite sums of Linear combination of a finite list, and the span span⁡(S) as the smallest linear subspace containing S read additively in the abelian group (V,+,0V).

[L1]

Finite sums: ∑i<0ui=0V; ∑i<σ(p)ui=(∑i<pui)+up; and the value depends only on u0,…,up−1 (The product g0g1⋯gn−1 of a finite list in a monoid, by recursion, with the empty product (n=0) equal to the identity, Linear combination of a finite list, and the span span⁡(S) as the smallest linear subspace containing S).

[L2]

(F1) a list all of whose entries are 0V sums to 0V; (F3) for j<p and u:p→V, ∑i<pui=uj+∑i<pui(j), where u(j) agrees with u at every i≠j and uj(j)=0V (The sum U+W of two linear subspaces and the sum ∑i<nUi of a finite family).

[L4]

(V,+,0V) is an abelian group; 0Fw=0V, λ0V=0V and (−1F)w=−w for all λ∈F, w∈V; 1Fw=w; and 1F≠0F in a field (Vector space over a field, In any vector space 0Fv=0V, λ0V=0V, (−λ)v=−(λv), (−1F)v=−v, and λv=0V forces λ=0F or v=0V, Field).

[L5]

A list v:p→V is linearly independent when ∑i<pλivi=0V forces λi=0F for every i<p, and a subset S⊆V is linearly independent when every injective finite list into S is (Linear independence: a finite list v:n→V is independent when ∑i<nλivi=0V forces every λi=0F, and a subset S⊆V is independent when every injective finite list into S is independent).

[L6]

Maps (Injection, surjection, bijection): a composite of injections is injective; a restriction of an injection is injective; an injection is a bijection onto its image and has a two-sided inverse there; and A≈B means a bijection A→B exists (Equinumerous sets, A≈B and A⪯B).

[L7]

Naturals (The natural numbers N (von Neumann), On N the order is membership: m<n  ⟺  m∈n, Order on the natural numbers): σ(p)=p∪{p} and p∉p; m<p  ⟺  m∈p and p={ m:m<p }; m<σ(p)  ⟺  m≤p; m<p  ⟺  σ(m)≤p (Discreteness: σ(n) is the immediate successor); exactly one of m<p, m=p, p<m holds (Trichotomy of the order on N); and every p≠0 is a successor (Every nonzero natural number is a successor).

[L8]

Splitting law for finite products in a monoid, read additively here: ∑i<a+qgi=(∑i<agi)+(∑j<qga+j) (Generalised associativity: in a monoid the product of a finite list does not depend on the bracketing, and in a commutative monoid it does not depend on the order of the factors either).

[L9]

Addition on N: m≤p means m+k=p for some k; ≤ is a total order; m+k<m+q  ⟺  k<q; and m+k=m+k′ forces k=k′ (Order on the natural numbers, ≤ is a linear order on N, Order is compatible with addition, Addition is cancellative, Addition is commutative).

Proof

technique · direct
1.1

Claim 1, by induction on m. At m=0 the image ι[0] is empty, so uj=0V for every j<n and (F1) gives ∑j<nuj=0V, which is also the empty sum ∑i<0uι(i). Assume the claim for m′, and let ι:σ(m′)→n be injective with u:n→V vanishing off ι[σ(m′)]. Put j0:=ι(m′)<n, let u(j0) be the list agreeing with u off j0 and equal to 0V at j0, and let ι′ be the restriction of ι to m′, which is injective. Then u(j0) vanishes off ι′[m′]: it vanishes at j0 by construction, and a j≠j0 outside ι′[m′] is outside ι[σ(m′)], so uj=0V. The inductive hypothesis therefore gives ∑j<nuj(j0)=∑i<m′uι(i)(j0)=∑i<m′uι(i), the second equality because ι(i)≠j0 for i<m′ by injectivity. Finally (F3) at j0 gives ∑j<nuj=uj0+∑j<nuj(j0)=∑i<m′uι(i)+uι(m′)=∑i<σ(m′)uι(i), by commutativity and the recursion.

L1L2L3L6L7
1.2

Claim 2, the deletion map. Fix n′∈N and k<σ(n′), so k≤n′. The two clauses define a function δk:n′→σ(n′): for i<k we have i<k≤n′, hence i<σ(n′), and for k≤i<n′ we have σ(i)≤n′<σ(n′). It is injective, being injective on each of the two blocks while its values on the first are below k and its values on the second satisfy k<σ(i). Its image is σ(n′)∖{k}: a j<σ(n′) with j<k is δk(j); a j<σ(n′) with k<j is nonzero, hence j=σ(i) for some i, and then i<j≤n′ gives i<n′ while k<σ(i) gives k≤i, so j=δk(i); and k itself is not a value, the first block giving values below k and the second values above k.

L7
1.3

Claim 3, the concatenated list. Let a,q∈N, y:a→V and z:q→V. Every i<a+q satisfies exactly one of i<a and a≤i; in the second case there is j with a+j=i, and a+j<a+q forces j<q, while j is unique by cancellation of addition. So the clauses ci:=yi for i<a and ca+j:=zj for j<q determine exactly one function c:a+q→V. If y and z are injective with y[a]∩z[q]=∅, then c is injective: it is injective on each block, and a value from the first block lies in y[a] while a value from the second lies in z[q], two disjoint sets. Its image is y[a]∪z[q] by the two clauses.

L6L7L9
1.4

Claim 3, the sum identity. The splitting law for finite sums gives ∑i<a+qci=(∑i<aci)+(∑j<qca+j), and by the defining clauses ci=yi for i<a and ca+j=zj for j<q, so ∑i<a+qci=∑i<ayi+∑j<qzj.

L1L8
1.5

Claim 4, injectivity. Let v:n→V be independent and suppose vj=vk with j≠k and j,k<n. Define λ:n→F by λj=1F, λk=−1F and λi=0F otherwise, and put ui:=λivi. Extracting the term at j by (F3) and then the term at k from the resulting list gives ∑i<nui=uj+(uk+∑i<nwi), where w agrees with u off {j,k} and is 0V at both. Every entry of w is 0V, since ui=0Fvi=0V for i∉{j,k}, so (F1) makes the last sum 0V. Hence ∑i<nλivi=1Fvj+(−1F)vk=vj+(−vk)=0V, while λj=1F≠0F, contradicting independence. So v is injective.

L1L2L4L5
1.6

Claim 4, no entry equal to 0V. Let v:n→V be independent and suppose vj=0V for some j<n. Define λ:n→F by λj=1F and λi=0F for i≠j, and put ui:=λivi, so ui=0Fvi=0V for every i≠j. Then (F3) at j together with (F1) gives ∑i<nλivi=uj+0V=1Fvj=vj=0V, while λj=1F≠0F, contradicting independence. So vi≠0V for every i<n.

L1L2L4L5
1.7

Zero extension of a list of scalars. Let ι:m→n be injective, let μ:m→F and let v:n→V. Because ι is injective there is exactly one λ:n→F with λι(i)=μi for every i<m and λj=0F for every j<n outside ι[m]. The list uj:=λjvj then satisfies uj=0Fvj=0V for every j outside ι[m], and uι(i)=μivι(i) for every i<m.

L4L6
1.8

Claim 7. Let S⊆V be independent and let T⊆S. Every injective finite list w:p→T is in particular an injective finite list into S, hence independent; so every injective finite list into T is independent, which is exactly independence of T.

L5
1.9

Claim 6, from right to left. Suppose v:n→V is injective and v[n] is an independent subset of V. Read as a function n→v[n], the list v is an injective finite list into v[n], hence independent; the vanishing condition is a condition on sums computed in V and is unaffected by which codomain v is read into, so the list v:n→V is independent. Moreover v is a bijection n→v[n], so v[n]≈n.

L5L6
2.1

Claim 2, the consequence. Let u:σ(n′)→V with uk=0V for some k<σ(n′). By step 1.2 the map δk is injective with image σ(n′)∖{k}, so u vanishes off that image, and claim 1, proved in step 1.1, gives ∑j<σ(n′)uj=∑i<n′uδk(i).

step 1.1step 1.2
2.2

Claim 5. Let v:n→V be independent, ι:m→n injective, and μ:m→F with ∑i<mμivι(i)=0V. Take the zero extension λ of step 1.7, so the list uj:=λjvj vanishes off ι[m] and uι(i)=μivι(i). Then step 1.1 gives ∑j<nλjvj=∑j<nuj=∑i<muι(i)=∑i<mμivι(i)=0V, so independence of v forces λj=0F for every j<n, and in particular μi=λι(i)=0F for every i<m. Hence v∘ι is independent.

step 1.1step 1.7L5
3.1

Claim 6, from left to right. Let v:n→V be independent; it is injective by step 1.5, so it is a bijection n→v[n] and has a two-sided inverse there. Let w:p→v[n] be an injective finite list; then ι:=v−1∘w:p→n is injective and v∘ι=w, so w is independent by step 2.2. Hence every injective finite list into v[n] is independent, that is, v[n] is an independent subset of V, and v[n]≈n.

step 1.5step 2.2L5L6
4.1

Claim 1 is step 1.1; claim 2 is step 1.2 with step 2.1; claim 3 is step 1.3 with step 1.4; claim 4 is step 1.5 with step 1.6; claim 5 is step 2.2; claim 6 is step 1.9 with step 3.1; and claim 7 is step 1.8.

step 1.1step 1.2step 1.3step 1.4step 1.5step 1.6step 1.8step 1.9step 2.1step 2.2step 3.1∎

Remarks

Depends on

Used by

Dependency tree · two levels

49 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