Scalable Nonlinear Learning with Adaptive Polynomial Expansions
Machine Learning
2014-10-03 v1 Machine Learning
Abstract
Can we effectively learn a nonlinear representation in time comparable to linear learning? We describe a new algorithm that explicitly and adaptively expands higher-order interaction features over base linear representations. The algorithm is designed for extreme computational efficiency, and an extensive experimental study shows that its computation/prediction tradeoff ability compares very favorably against strong baselines.
Cite
@article{arxiv.1410.0440,
title = {Scalable Nonlinear Learning with Adaptive Polynomial Expansions},
author = {Alekh Agarwal and Alina Beygelzimer and Daniel Hsu and John Langford and Matus Telgarsky},
journal= {arXiv preprint arXiv:1410.0440},
year = {2014}
}
Comments
To appear in NIPS 2014