Revisiting Superlinear Convergence of Proximal Newton-Like Methods to Degenerate Solutions
Ching-pei Lee, Stephen J. Wright · 2026
Read Revisiting Superlinear Convergence of Proximal Newton-Like Methods to Degenerate Solutions by Ching-pei Lee, Stephen J. Wright (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-10470.txt.
Description
Ching-pei Lee and Stephen J. Wright revisit fast settling for a class of tuning methods in a 2026 paper. Proximal Newton-like methods solve regularized problems, but their superlinear settling was understood mainly for well-behaved cases. Broken answers, where Jacobians turn single, had weaker promises.
The authors show superlinear settling can survive degeneracy under a Holderian error bound. This holds even when the map is nonmonotone and its Jacobian is merely uniformly steady rather than Lipschitz. For convex regularized problems they add a globalization plan that promises strict objective decrease and avoids the Maratos effect. The results widen the class of problems where builders can expect rapid local settling.
Catalog record
| CREATOR | Ching-pei Lee, Stephen J. Wright |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 81 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-10470