Distributed Multi-Task Learning with Shared Representation
Machine Learning
2016-03-08 v1 Machine Learning
Abstract
We study the problem of distributed multi-task learning with shared representation, where each machine aims to learn a separate, but related, task in an unknown shared low-dimensional subspaces, i.e. when the predictor matrix has low rank. We consider a setting where each task is handled by a different machine, with samples for the task available locally on the machine, and study communication-efficient methods for exploiting the shared structure.
Cite
@article{arxiv.1603.02185,
title = {Distributed Multi-Task Learning with Shared Representation},
author = {Jialei Wang and Mladen Kolar and Nathan Srebro},
journal= {arXiv preprint arXiv:1603.02185},
year = {2016}
}