Reinterpreting EMML as Mirror Descent for Constrained Maximum Likelihood Estimation
Antonin Clerc, Ségolène Martin, Nicolas Papadakis, Gabriele Steidl, [0.5em] Institute of Mathematics, Technische Universität Berlin, Germany Univ. Bordeaux, CNRS, Bordeaux INP, IMB, UMR 5251, F-33400 Talence, France [email protected], [email protected] [email protected] · 2026
Read Reinterpreting EMML as Mirror Descent for Constrained Maximum Likelihood Estimation by Antonin Clerc, Ségolène Martin, Nicolas Papadakis, Gabriele Steidl, [0.5em] Institute of Mathematics, Technische Universität Berlin, Germany Univ. Bordeaux, CNRS, Bordeaux INP, IMB, UMR 5251, F-33400 Talence, France [email protected], [email protected] [email protected] (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-2602-13063.txt.
Description
The EMML recipe reconstructs images under Poisson noise and belongs to the expectation-maximisation family, but its deeper structure has been unclear. Antonin Clerc and three co-authors reinterpret EMML as mirror descent applied to a reparametrised objective. That viewpoint lets convex constraints enter through Bregman projections while preserving the multiplicative update structure that keeps EMML lean.
The authors then prove the limited recipe converges to a answer of the limited maximum-odds task. Number tests on hyperspectral unmixing show the limited version converging in fewer iterations than standard EMML. For image reconstruction, the reinterpretation is both conceptual and practical: an old recipe gains constraints and a convergence story through a modern tuning lens.
Catalog record
| CREATOR | Antonin Clerc, Ségolène Martin, Nicolas Papadakis, Gabriele Steidl, [0.5em] Institute of Mathematics, Technische Universität Berlin, Germany Univ. Bordeaux, CNRS, Bordeaux INP, IMB, UMR 5251, F-33400 Talence, France [email protected], [email protected] [email protected] |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 20 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-13063