Warm Start for Parameter Selection of Linear Classifiers
Date Issued
2015
Date
2015
Author(s)
Chu, Bo-Yu
Abstract
In linear classification, a regularization term effectively remedies the overfitting problem, but selecting a good regularization parameter is usually time consuming. We consider cross validation for the selection process, so severaloptimization problems under different parameters must be solved. Our aim is to devise effective warm-start strategies to efficiently solve this sequence of optimization problems. We detailedly investigate the relationship between optimalsolutions of logistic regression/linear SVM and regularization parameters. Based on the analysis, we develop an efficient tool to automatically find a suitable parameter for users with no related background knowledge.
Subjects
warm start, regularization
parameter
parameter selection
linear classification
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-104-R02222047-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):5e18133918a6ab1ff61d642ca7796095