Bregman Distance to L1 Regularized Logistic Regression - Computer Science > LearningReport as inadecuate




Bregman Distance to L1 Regularized Logistic Regression - Computer Science > Learning - Download this document for free, or read online. Document in PDF available to download.

Abstract: In this work we investigate the relationship between Bregman distances andregularized Logistic Regression model. We present a detailed study of BregmanDistance minimization, a family of generalized entropy measures associated withconvex functions. We convert the L1-regularized logistic regression into thismore general framework and propose a primal-dual method based algorithm forlearning the parameters. We pose L1-regularized logistic regression intoBregman distance minimization and then apply non-linear constrainedoptimization techniques to estimate the parameters of the logistic model.



Author: Mithun Das Gupta, Thomas S. Huang

Source: https://arxiv.org/







Related documents