Matej Balog, Balaji Lakshminarayanan, Zoubin Ghahramani, Daniel M. Roy, Yee Whye Teh
Thirty-Second Conference on Uncertainty in Artificial Intelligence (UAI), 2016.
[PDF] [supp] [arXiv] [poster] [slides]
The scripts provided here implement experiments from this paper. The scripts experiment_1_laplace_kernel_approximation, experiment_2_fast_kernel_width_learning and experiment_3_mondrian_kernel_vs_forest are intended to be directly runnable.
Python packages: heapq, matplotlib, numpy, scipy, sklearn, sys, time
The CPU dataset cpu.mat can be download and extracted from here.
A bug in scipy may cause the Python kernel to restart when loading the CPU dataset from cpu.mat. Downgrading to scipy 0.16.0 should solve the problem.
@inproceedings{balog2016mondriankernel,
author = {Matej Balog and Balaji Lakshminarayanan and Zoubin Ghahramani and Daniel M.~Roy and Yee Whye Teh},
title={The {M}ondrian Kernel},
booktitle = {32nd Conference on Uncertainty in Artificial Intelligence (UAI)},
year = {2016},
month = {June},
url = {http://www.auai.org/uai2016/proceedings/papers/236.pdf}
}