arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.07258cs.LGstat.ML
收藏

利用安全筛选加速非负和有界变量线性回归算法

Accelerating Non-Negative and Bounded-Variable Linear Regression Algorithms with Safe Screening

INRAE(法国国家农业食品与环境研究院);CNRS(法国国家科学研究中心);University of Toulouse(图卢兹大学)

Cassio F. Dantas, Emmanuel Soubies, Cédric Févotte

AI总结:

本文提出一种基于安全筛选的加速技术,通过在迭代中识别饱和坐标来加速非负和有界变量线性回归求解器,并提供理论保证和实验验证。

摘要:

非负和有界变量线性回归问题在机器学习和信号处理中有广泛应用。本文提出一种技术,通过在迭代过程中识别饱和坐标来加速现有求解器。这类似于先前为稀疏正则化回归问题提出的安全筛选技术。该策略具有可证明的安全性,因为它提供了理论保证,确保被识别的坐标在最优解中确实饱和。在合成和真实数据上的实验结果表明,非负和有界变量问题均获得了显著的加速效果。

英文摘要:

Non-negative and bounded-variable linear regression problems arise in a variety of applications in machine learning and signal processing. In this paper, we propose a technique to accelerate existing solvers for these problems by identifying saturated coordinates in the course of iterations. This is akin to safe screening techniques previously proposed for sparsity-regularized regression problems. The proposed strategy is provably safe as it provides theoretical guarantees that the identified coordinates are indeed saturated in the optimal solution. Experimental results on synthetic and real data show compelling accelerations for both non-negative and bounded-variable problems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑