Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passverified 2026-10-23 (gpt-6.1-sol)
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.

Continuation of bounded polynomial ODE solutions

Statement

Let N≥1, let F:RN→RN be a polynomial map, and let T∗>0. If x∈C1([0,T∗),RN) solves x′(t)=F(x(t)) and there is R≥0 such that ∥x(t)∥2≤R for every 0≤t<T∗, then there are ϵ>0 and a solution x~∈C1([0,T∗+ϵ),RN) of the same ODE with x~(t)=x(t) for 0≤t<T∗. Thus a finite right endpoint of a bounded finite-dimensional polynomial ODE trajectory is extendible.

Facts & Assumptions

Given: The data in the Statement, with x(0)=x0∈RN.

[F1]

If F is continuous and locally Lipschitz in its state variable on an open domain, Picard–Lindelöf gives a unique local solution through every point of that domain (Picard-Lindelöf local existence and uniqueness for first-order systems).

[F3]

If a continuous curve f:[a,b]→RN is differentiable on (a,b) and ∥f′(t)∥2≤M, then ∥f(t)−f(s)∥2≤M∣t−s∣ for all s,t∈[a,b] (If f:[a,b]→Rm is differentiable with integrable f′ then ∫abf′=f(b)−f(a); and a bounded derivative makes f Lipschitz).

[F4]

A continuous curve solves an IVP exactly when it satisfies the associated oriented Volterra integral equation (A first-order initial value problem is equivalent to its Volterra integral equation).

Proof

technique · direct polynomial bound and endpoint continuation
1.1givenalgebraF1

Write each component as a finite sum Fi(x)=∑α∈Eiciαxα. On any bounded coordinate box, for each exponent q≥1 the identity aq−bq=(a−b)∑j=0q−1aq−1−jbj bounds the difference of each monomial at two points by a finite constant times their Euclidean distance; telescoping over the N factors and summing the finitely many monomials proves that F is locally Lipschitz there. It is also continuous, so [F1] applies at every state.

2.1givenstep 1.1algebra

Put R+=max⁡(1,R) and MR=(∑i=1N(∑α∈Ei∣ciα∣R+∣α∣)2)1/2. Since ∣xj(t)∣≤R for each coordinate, the finite polynomial sums give ∥F(x(t))∥2≤MR for all t<T∗.

3.1givenF2F3step 2.1

For 0≤s<t<T∗, apply [F3] to x∣[s,t] to obtain ∥x(t)−x(s)∥2≤MR(t−s). Thus, for tn=T∗(1−1/n), n≥2, each coordinate sequence (xi(tn))n≥2 is Cauchy because ∣xi(tn)−xi(tm)∣≤∥x(tn)−x(tm)∥2≤MR∣tn−tm∣. By [F2], each coordinate has a real limit ξi; set x∗=(ξ1,…,ξN)∈RN. For any ε>0, choose for each of the finitely many coordinates an index after which ∣xi(tn)−ξi∣<ε/N, and take the maximum of those indices. Then ∥x(tn)−x∗∥22=∑i=1N∣xi(tn)−ξi∣2<ε2, so x(tn)→x∗. For any t<T∗ and all sufficiently large n with tn≥t, the same Lipschitz bound gives ∥x(t)−x∗∥2≤MR(tn−t)+∥x(tn)−x∗∥2; letting n→∞ yields ∥x(t)−x∗∥2≤MR(T∗−t), so x(t)→x∗. The closed-ball bound passes to the limit, hence ∥x∗∥2≤R.

4.1step 1.1step 3.1F1

By step 1.1 the polynomial field is locally Lipschitz near x∗. Apply [F1] to the IVP at (T∗,x∗) to obtain a local solution y on an open interval about T∗.

5.1step 3.1step 4.1F4

By [F4], x(t)=x0+∫0tF(x(s)) ds for t<T∗ and y(t)=x∗+∫T∗tF(y(s)) ds near T∗. The limit in step 3.1 and continuity of F extend the first identity to t=T∗. Define x~=x before T∗ and x~=y from T∗ onward, assigning the common value x∗ at T∗. The joined curve is continuous and the two integral identities add to its Volterra equation on [0,T∗+ϵ) for some ϵ>0; [F4] therefore makes it a C1 solution of the same ODE extending x.

6.1step 5.1∎

The extension in step 5.1 agrees with the original curve at every t<T∗ and is defined beyond T∗. This proves the statement.

Depends on

Dependency tree · 2 levels

Sources