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.
Two initial intervals converge to different minima of a periodic objective
Example
Let on , and use the fixed step . Its gradient is , which is -Lipschitz, and . Gradient descent is the iteration . Every start produces a sequence converging to the global minimizer , while every start produces a sequence converging to the distinct global minimizer .
Facts & Assumptions
Given: The objective , the fixed step , and an initial point in either or .
Gradient descent: with a fixed step , the update is .
Gradient descent has vanishing gradients when the objective is bounded below along its iterates: for a continuously differentiable objective with -Lipschitz gradient, fixed , and an objective lower bounded along the iterates, the gradient norms tend to zero.
Sums, scalar multiples, products and quotients: , , , and when : derivatives of sums and scalar multiples are the corresponding sums and scalar multiples of derivatives, and constant functions have derivative zero.
The derivative of at a point that is a limit point of , and differentiability on a set: is the limit of as through ; a constant function has zero difference quotient.
Sine and cosine are -Lipschitz on : and for all real .
The zero sets of sine and cosine and the least positive common period 2 pi: exactly when for an integer , and both sine and cosine have period .
A monotone sequence converges if and only if it is bounded: every bounded monotone real sequence converges.
Verification
By [F3, F4, F5], . Then [F6] makes the gradient continuous and -Lipschitz; [F9] gives , and [F3, F8] give . Thus the hypotheses of [F2] hold with and , so every trajectory has gradient norm tending to zero.
Suppose . If , [F7] gives , and [F6] with gives ; hence . If , it stays at . Induction shows , so [F10] gives a limit . Given , convergence gives an index after which both and ; hence and, by the update, . Also [F6] gives , so . By [F8], the only such zero in is , so .
Suppose . If , put . By [F7, F8, F9], , and [F6] gives . Therefore ; if , it remains there. Induction gives a nondecreasing sequence bounded above by , so [F10] gives a limit . Convergence implies by the same two-term estimate used in step 1.2, so . Since , [F6] gives , hence . By [F8], the only such zero with is . Thus .
Depends on
- Gradient descent has vanishing gradients when the objective is bounded below along its iterates
- Gradient descent
- The Jacobian matrix of partial derivatives and the gradient in the scalar-valued case
- The derivatives of sine and cosine are cosine and minus sine
- Sums, scalar multiples, products and quotients: $(f+g)'(c) = f'(c) + g'(c)$, $(\alpha f)'(c) = \alpha f'(c)$, $(fg)'(c) = f'(c)g(c) + f(c)g'(c)$, and $(f/g)'(c) = \bigl(f'(c)g(c) - f(c)g'(c)\bigr)/g(c)^{2}$ when $g(c) \ne 0$
- The derivative $f'(c) = \lim_{x \to c} \frac{f(x) - f(c)}{x - c}$ of $f : A \to \mathbb{R}$ at a point $c \in A$ that is a limit point of $A$, and differentiability on a set
- Sine and cosine are $1$-Lipschitz on $\mathbb{R}$
- Pi is the first positive zero of sine
- The zero sets of sine and cosine and the least positive common period 2 pi
- Parity and the Pythagorean identity for sine and cosine
- A monotone sequence converges if and only if it is bounded
Used by
Nothing in the library uses this result yet.
Dependency tree · two levels
51 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.