Entropy Adjusted Graphical Lasso for Sparse Precision Matrix Estimation
Vahe Avagyan · 2025
Read Entropy Adjusted Graphical Lasso for Sparse Precision Matrix Estimation by Vahe Avagyan (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-2501-05308.txt.
Description
Vahe Avagyan adds an entropy term to the graphical lasso in a 2025 paper. A precision matrix sums up how variables link once the rest are held fixed. With many variables, the matrix must be fit with a penalty.
The usual penalties shrink entries toward zero and give a sparse, stable fit. But they do not deal with how unsure that fit is. A sparser matrix leaves the spread wider, and so its entropy is higher. The new method adds a log-determinant term to the penalty to price that doubt. Theory and examples show how this trade-off shifts. Readers in this field get a tool that weighs its own doubt, not just how sparse it is.
Catalog record
| CREATOR | Vahe Avagyan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 78 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-05308