AISTATS 2018poster0 citations
Teacher Improves Learning by Selecting a Training Subset
Yuzhe Ma, Robert Nowak, Philippe Rigollet, Xuezhou Zhang, Xiaojin Zhu
Abstract
We call a learner super-teachable if a teacher can trim down an iid training set while making the learner learn even better. We provide sharp super-teaching guarantees on two learners: the maximum likelihood estimator for the mean of a Gaussian, and the large margin classifier in 1D. For general learners, we provide a mixed-integer nonlinear programming-based algorithm to find a super teaching set. Empirical experiments show that our algorithm is able to find good super-teaching sets for both regression and classification problems.
BibTeX
@InProceedings{pmlr-v84-ma18a,
title = {Teacher Improves Learning by Selecting a Training Subset},
author = {Ma, Yuzhe and Nowak, Robert and Rigollet, Philippe and Zhang, Xuezhou and Zhu, Xiaojin},
booktitle = {Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics},
pages = {1366--1375},
year = {2018},
editor = {Storkey, Amos and Perez-Cruz, Fernando},
volume = {84},
series = {Proceedings of Machine Learning Research},
month = {09--11 Apr},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v84/ma18a/ma18a.pdf},
url = {https://proceedings.mlr.press/v84/ma18a.html},
abstract = {We call a learner super-teachable if a teacher can trim down an iid training set while making the learner learn even better. We provide sharp super-teaching guarantees on two learners: the maximum likelihood estimator for the mean of a Gaussian, and the large margin classifier in 1D. For general learners, we provide a mixed-integer nonlinear programming-based algorithm to find a super teaching set. Empirical experiments show that our algorithm is able to find good super-teaching sets for both regression and classification problems. }
}