On the Depth Transfer of Neural Optimizers via L-Smoothness ZVCD8RG1JJ
| URI | m2://res/ZVCD8RG1JJ |
|---|---|
| Kind | preprint |
| Creators | Maximilian Wong, Chenyu Xu, Jiajin Li, Yiping Lu |
| Date | 2026-10-09 |
| Subjects | arxiv cs.LG, arxiv math.OC |
| Scope | border |
Expressions
e1: version, preprint
| Date | 2026-10-09 |
|---|---|
| Ids | arXiv 2610.14344 |
Items in this resource
The mentions of the inventory that name this resource. Each name links to the card of its inventory row.
| Item | Kind | Where | Source |
|---|---|---|---|
| Derivative Bounds Implied by Assumption 2 | statement | corollary 4 | arxiv |
| Attention blocks satisfy the residual regularity | statement | proposition 2 | arxiv |
| Directional Derivatives | definition | definition 2 | arxiv |
| Directional Hessian Smoothness Bound | statement | lemma 2 | arxiv |
| Directional Derivative Lipschitz Bound | statement | lemma 1 | arxiv |
| Jacobian Bound | statement | corollary 2 | arxiv |
| Depth-Independent \(L\)-Smoothness | statement | theorem 2 | arxiv |
| Depth-Independent Lipschitz Continuity | statement | theorem 1 | arxiv |
| Uniform Bound on First-Order Directional Derivatives | statement | corollary 5 | arxiv |
| Jacobian Product Bound | statement | corollary 3 | arxiv |
| Abstract Residual Network | definition | definition 1 | arxiv |
| MLP residual satisfies Assumption 2 | statement | proposition 1 | arxiv |
| MLP residual networks satisfy the main theorems | statement | corollary 6 | arxiv |
| Forward Boundedness | statement | corollary 1 | arxiv |
Catalogue records
- arxiv: 2610.14344 (work)