Improvement of the SAM algorithm based on causal structure analysis

In response to the computational complexity issue of the directed acyclic graph constraint in the Structure-Agnostic Model (SAM), we have proposed an innovative solution: an improved algorithm for the directed acyclic graph constraint based on continuous optimization. This algorithm, while maintaini...

Full description

Saved in:
Bibliographic Details
Published inProceedings of SPIE, the international society for optical engineering Vol. 13486; pp. 134862P - 134862P-8
Main Authors Jia, Bin, Cai, Qingsong, Jia, Ming
Format Conference Proceeding
LanguageEnglish
Published SPIE 09.01.2025
Online AccessGet full text
ISBN1510687610
9781510687615
ISSN0277-786X
DOI10.1117/12.3055811

Cover

More Information
Summary:In response to the computational complexity issue of the directed acyclic graph constraint in the Structure-Agnostic Model (SAM), we have proposed an innovative solution: an improved algorithm for the directed acyclic graph constraint based on continuous optimization. This algorithm, while maintaining the original acyclic constraint conditions, ingeniously introduces a hyperparameter α as an important component of the constraint. This improvement not only effectively reduces the computational complexity but also enhances the precision of the calculations to a certain extent.
Bibliography:Conference Date: 2024-08-30|2024-09-01
Conference Location: Chengdu, China
ISBN:1510687610
9781510687615
ISSN:0277-786X
DOI:10.1117/12.3055811