Introduction

ftrl is a Java version of Follow-the-Regularized-Leader algorithm published in paper "Follow-the-Regularized-Leader and Mirror Descent: Equivalence Theorems and L1 Regularization". It can be used for classification problems with online convex optimization.

Usage

java -cp *.jar com.ilighti.ml.app.Train data/sample_libsvm_data.txt model

java -cp *.jar com.ilighti.ml.app.Predict data/sample_libsvm_data.txt model ouput

Bug report

Contact [email protected] to report any bugs.

Liecence

The author is Xiaoyu Fang (Rain) from China. Please quot the source whenever you use it.

Reference

[1]Follow-the-Regularized-Leader and Mirror Descent: Equivalence Theorems and L1 Regularization. H. Brendan McMahan

[2]Ad Click Prediction: a View from the Trenches. H. Brendan McMahan, Gary Holt, D. Sculley, Michael Young,Dietmar Ebner, Julian Grady, Lan Nie, Todd Phillips, Eugene Davydov,Daniel Golovin, Sharat Chikkerur, Dan Liu, Martin Wattenberg, Arnar Mar Hrafnkelsson, Tom Boulos, Jeremy Kubica