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.
Local Gram positivity and loss smoothness give confined gradient descent
Statement
Let be , , and . On the closed ball suppose a block of the output Jacobian satisfies and , with . If and , every fixed-step gradient-descent iterate stays in the ball,
The assertion is trivial for ; otherwise the hypotheses imply .
Statement references
Facts & Assumptions
Given: The output map, quadratic loss, ball, spectral and Hessian bounds, and step size.
Gradient descent uses . Gradient descent
For a scalar function on a segment, the first-order Taylor remainder is the integral of its second derivative against the linear remainder kernel. Taylor's theorem with an integral remainder
The Hessian norm is the induced operator norm on the finite-dimensional Euclidean parameter space. The operator norm is zero on the zero domain and otherwise is max_{||v||=1} ||Tv||
Euclidean Cauchy–Schwarz gives for all real vectors of the same finite dimension. Cauchy-Schwarz with its equality case, the triangle inequality for , the parallelogram law and polarisation
Proof
Write . Since , in the ball. For any step , , whose segment stays in the ball, put for . Then and , where . By Euclidean Cauchy–Schwarz [F4] and the induced norm [F3], ; for the same bound is immediate. By [F2], . If , nonnegativity of gives .
For such a segment with , . If , and hence , so the step is stationary. Telescoping any finite sequence of admissible full or truncated steps gives total parameter path length at most .
Suppose there is a first proposed full step exiting the ball. Truncate that step at the first intersection with its boundary; this is a segment of length with , wholly in the ball. Step 2.1 applies to all previous steps and this truncated step, giving total path length strictly less than . Reaching the boundary from requires length at least , a contradiction. Thus no exit occurs. Step 1.1 applied at every full step gives the loss rate, and taking the supremum of finite partial path lengths in step 2.1 gives the asserted series bound.
Depends on
- Gradient descent
- Real spectral theorem: a self-adjoint endomorphism of a finite-dimensional real inner product space has an orthonormal eigenbasis
- Taylor's theorem with an integral remainder
- The operator norm is zero on the zero domain and otherwise is max_{||v||=1} ||Tv||
- Cauchy-Schwarz $\lvert\langle x,y\rangle\rvert \le \lVert x\rVert_2\lVert y\rVert_2$ with its equality case, the triangle inequality for $\lVert\cdot\rVert_2$, the parallelogram law and polarisation
Used by
Dependency tree · two levels
28 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
- Du, Lee, Li, Wang and Zhai, Gradient Descent Finds Global Minima of Deep Neural Networks, Appendices B, C, E and G (standard reference, not scraped)