AOSO-LogitBoost: Adaptive One-Vs-One LogitBoost for Multi-Class Problem

Details AOSO-LogitBoost: Adaptive One-Vs-One LogitBoost for Multi-Class Problem AOSO-LogitBoost: Adaptive One-Vs-One LogitBoost for Multi-Class Problem

2011-10-18

arxiv.org/abs/1110.3907v1

Statistics

Machine Learning

Scientific paper

LogitBoost and its later improvement, ABC-LogitBoost, are both successful multi-class boosting algorithms for classification. In this paper, we explicitly formulate the tree building at each LogitBoost iteration as constrained quadratic optimization. Both LogitBoost and ABC-LogtiBoost adopt approximated solver to such quadratic subproblem. We then propose an intuitively more natural solver, i.e. the block coordinate descent algorithm, and demonstrate that it leads to higher classification accuracy and faster convergence rate on a number of public datasets. This new LogitBoost behaves as if it combines many one-vs-one binary classifiers adaptively, hence the name AOSO-LogitBoost(Adaptive One-vs-One LogitBoost)

Affiliated with

Also associated with

No associations

Rating

AOSO-LogitBoost: Adaptive One-Vs-One LogitBoost for Multi-Class Problem does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with AOSO-LogitBoost: Adaptive One-Vs-One LogitBoost for Multi-Class Problem, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and AOSO-LogitBoost: Adaptive One-Vs-One LogitBoost for Multi-Class Problem will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-290521