Universality of first-order methods on random and deterministic matrices
Nicola Gorini, Chris Jones, Dmitriy Kunisky, Lucas Pesenti · 2026
Read Universality of first-order methods on random and deterministic matrices by Nicola Gorini, Chris Jones, Dmitriy Kunisky, Lucas Pesenti (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-2604-11729.txt.
Introduction
Gorini, Jones, Kunisky, and Pesenti analyse first-order methods, algorithms updating a state vector by matrix-vector products and entrywise nonlinearities. Via diagram expansions and the traffic distribution of the input matrix, they compute that distribution for the first non-trivial deterministic matrices, including Walsh-Hadamard and discrete sine and cosine transforms. That settles parts of longstanding conjectures from Marinari, Parisi, and Ritort. They also design a new approximate message passing iteration whose dynamics stay Gaussian given latent random variables, with a counting-based reading of the Onsager correction.
Catalog record
| CREATOR | Nicola Gorini, Chris Jones, Dmitriy Kunisky, Lucas Pesenti |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 307790 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.11729