Continuous Kernel Learning
Abstract
Kernel learning is the problem of determining the best kernel (either from a dictionary of fixed kernels, or from a smooth space of kernel representations) for a given task. In this paper, we describe a new approach to kernel learning that establishes connections between the Fourier-analytic representation of kernels arising out of Bochner’s theorem and a specific kind of feed-forward network using cosine activations. We analyze the complexity of this space of hypotheses and demonstrate empirically that our approach provides scalable kernel learning superior in quality to prior approaches.
Cite
Text
Moeller et al. "Continuous Kernel Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2016. doi:10.1007/978-3-319-46227-1_41Markdown
[Moeller et al. "Continuous Kernel Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2016.](https://mlanthology.org/ecmlpkdd/2016/moeller2016ecmlpkdd-continuous/) doi:10.1007/978-3-319-46227-1_41BibTeX
@inproceedings{moeller2016ecmlpkdd-continuous,
title = {{Continuous Kernel Learning}},
author = {Moeller, John and Srikumar, Vivek and Swaminathan, Sarathkrishna and Venkatasubramanian, Suresh and Webb, Dustin},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2016},
pages = {657-673},
doi = {10.1007/978-3-319-46227-1_41},
url = {https://mlanthology.org/ecmlpkdd/2016/moeller2016ecmlpkdd-continuous/}
}