The Mondrian Kernel
arXiv (Cornell University)2016pp. 32–41
Citations Over TimeTop 12% of 2016 papers
Abstract
We introduce the Mondrian kernel, a fast $\textit{random feature}$ approximation to the Laplace kernel. It is suitable for both batch and online learning, and admits a fast kernel-width-selection procedure as the random features can be re-used efficiently for all kernel widths. The features are constructed by sampling trees via a Mondrian process [Roy and Teh, 2009], and we highlight the connection to Mondrian forests [Lakshminarayanan et al., 2014], where trees are also sampled via a Mondrian process, but fit independently. This link provides a new insight into the relationship between kernel methods and random forests.
Related Papers
- → Kernel Optimization in Discriminant Analysis(2010)95 cited
- → Exploring different kernel functions for kernel-based clustering(2016)3 cited
- → Theoretical properties and implementation of the one-sided mean kernel for time series(2015)
- → Variance analyses for kernel regressors with nested reproducing kernel hilbert spaces(2012)