Permanents of matrix ensembles: computation, distribution, and geometry
Igor Rivin · 2026
Read Permanents of matrix ensembles: computation, distribution, and geometry by Igor Rivin (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-2602-10141.txt.
Description
Permanents, the unsigned cousins of determinants, are known to be hard to compute, yet their distributions reveal deep structure. A 2026 paper by Igor Rivin reports a computational and experimental study using graphics hardware to speed permanent computation across complex, real, finite-field, and rational matrices. For Haar-distributed unitary matrices, the permanent follows a circularly symmetric complex Gaussian law, confirmed with tens of thousands of samples for sizes up to twenty-three.
For orthogonal matrices, the permanent looks nearly real Gaussian with excess kurtosis fading slowly. For Gaussian ensembles, it follows a heavy-tailed stable law well below Gaussian stability, and a conjecture about log-shaped tails holds for some ensembles but fails for others. Along geodesics on the unitary group, a shared scaling function emerges independent of size.
Catalog record
| CREATOR | Igor Rivin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-10141