Training-Time Optimization of a Budgeted Booster
Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence. Special Track on Machine Learning. Pages 3583-3588.
Abstract
We consider the problem of feature-efficient prediction - a setting where features have costs and the learner is limited by a budget constraint on the total cost of the features it can examine in test time. We focus on solving this problem with boosting by optimizing the choice of base learners in the training phase and stopping the boosting process when the learner's budget runs out. We experimentally show that our method improves upon the boosting approach AdaBoostRS [Reyzin, 2011] and in many cases also outperforms the recent algorithm SpeedBoost [Grubb and Bagnell, 2012]. We provide a theoretical justication for our optimization method via the margin bound. We also experimentally show that our method outperforms pruned decision trees, a natural budgeted classifier.
@inproceedings{ijcai2015p504,
title = {Training-Time Optimization of a Budgeted Booster},
author = {Huang, Yi and Powers, Brian and Reyzin, Lev},
booktitle = {Proceedings of the Twenty-Fourth International Joint Conference on
Artificial Intelligence, {IJCAI-15}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Qiang Yang and Michael Wooldridge},
pages = {3583--3588},
year = {2015},
month = {7},
note = {Special Track on Machine Learning},
url = {https://ijcai.org/proceedings/2015/504},
}
TY - CPAPER TI - Training-Time Optimization of a Budgeted Booster AU - Huang, Yi AU - Powers, Brian AU - Reyzin, Lev T2 - Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-15 PB - International Joint Conferences on Artificial Intelligence PY - 2015 SP - 3583 EP - 3588 UR - https://ijcai.org/proceedings/2015/504 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
