Relocated Fixed-Point Iterations with Applications to Variable Stepsize Resolvent Splitting
Felipe Atenas, Heinz H. Bauschke, Minh N. Dao, Matthew K. Tam · 2025
Read Relocated Fixed-Point Iterations with Applications to Variable Stepsize Resolvent Splitting by Felipe Atenas, Heinz H. Bauschke, Minh N. Dao, Matthew K. Tam (2025) 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-2507-07428.txt.
Description
Felipe Atenas and three co-authors build a convergence theory for algorithms whose target moves at every step. Their work appeared in twenty twenty-five. Many optimization routines chase a fixed point, but real methods often swap operators as they go, leaving no single point that fits all steps.
Their trick adds a relocation step after each main update, carrying the current fixed point onto the next operator's ground. A widened version of a classic demiclosedness principle powers the proof. The payoff is a graph-based Douglas-Rachford method for splitting problems into many operators. Its step size may now vary between iterations, a freedom earlier versions lacked. The framework widens the design space for splitting algorithms generally.
Catalog record
| CREATOR | Felipe Atenas, Heinz H. Bauschke, Minh N. Dao, Matthew K. Tam |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 90 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-07428