Aspects of a Generalized Theory of Sparsity based Inference in Linear Inverse Problems
Ryan O’Dowd 1 2, Raghu G. Raj 1, Hrushikesh N. Mhaskar 2 · 2025
Read Aspects of a Generalized Theory of Sparsity based Inference in Linear Inverse Problems by Ryan O’Dowd 1 2, Raghu G. Raj 1, Hrushikesh N. Mhaskar 2 (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-2503-00178.txt.
Description
Raghu Raj, Ryan O'Dowd, Owen Huber, and Hrushikesh Mhaskar lay groundwork, in a paper published in 2025, for a broader theory of sparse recovery. Sparse methods rebuild a signal from few measures by assuming most of its parts are zero. They power compressive sensing, which samples far less than older rules demand. A newer family based on mixed bell-shaped noise priors often beats them in practice, but lacked a theory.
The authors define a wider notion of sparsity that can hold both families. Within it, they state a key condition on the sensing setup and adapt two classic algorithms, basis pursuit and a reweighted squares method. They show a subset of the newer priors fits the frame. The work aims to explain why the newer methods win.
Catalog record
| CREATOR | Ryan O’Dowd 1 2, Raghu G. Raj 1, Hrushikesh N. Mhaskar 2 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-00178