Not-So-Random Features
Machine Learning
2018-02-28 v2 Machine Learning
Abstract
We propose a principled method for kernel learning, which relies on a Fourier-analytic characterization of translation-invariant or rotation-invariant kernels. Our method produces a sequence of feature maps, iteratively refining the SVM margin. We provide rigorous guarantees for optimality and generalization, interpreting our algorithm as online equilibrium-finding dynamics in a certain two-player min-max game. Evaluations on synthetic and real-world datasets demonstrate scalability and consistent improvements over related random features-based methods.
Cite
@article{arxiv.1710.10230,
title = {Not-So-Random Features},
author = {Brian Bullins and Cyril Zhang and Yi Zhang},
journal= {arXiv preprint arXiv:1710.10230},
year = {2018}
}
Comments
Published as a conference paper at ICLR 2018