This paper concerns the distributed training of nonlinear kernel machines on Map-Reduce. We show that a re-formulation of Nystr\"om approximation based solution which is solved using gradient based techniques is well suited for this, especially when it is necessary to work with a large number of basis points. The main advantages of this approach are: avoidance of computing the pseudo-inverse of the kernel sub-matrix corresponding to the basis points; simplicity and efficiency of the distributed part of the computations; and, friendliness to stage-wise addition of basis points. We implement the method using an AllReduce tree on Hadoop and demonstrate its value on a few large benchmark datasets.
@article{arxiv.1405.4543,
title = {A Distributed Algorithm for Training Nonlinear Kernel Machines},
author = {Dhruv Mahajan and S. Sathiya Keerthi and S. Sundararajan},
journal= {arXiv preprint arXiv:1405.4543},
year = {2014}
}