Publication Type

Journal Article

Version

publishedVersion

Publication Date

4-2018

Abstract

Although dispersing one single task to distributed learning nodes has been intensively studied by the previous research, multi-task learning on distributed networks is still an area that has not been fully exploited, especially under decentralized settings. The challenge lies in the fact that different tasks may have different optimal learning weights while communication through the distributed network forces all tasks to converge to an unique classifier. In this paper, we present a novel algorithm to overcome this challenge and enable learning multiple tasks simultaneously on a decentralized distributed network. Specifically, the learning framework can be separated into two phases: (i) multi-task information is shared within each node on the first phase; (ii) communication between nodes then leads the whole network to converge to a common minimizer. Theoretical analysis indicates that our algorithm achieves a (Formula presented.) regret bound when compared with the best classifier in hindsight, which is further validated by experiments on both synthetic and real-world datasets.

Keywords

Decentralized distributed learning, Multi-task learning, Online learning, Classification (of information) Learning systems, Distributed learning, Distributed networks, Learning frameworks, Multiple tasks, Multitask learning, Novel algorithm, Online learning, Real-world datasets, E-learning

Discipline

Computer Sciences | Databases and Information Systems | Theory and Algorithms

Research Areas

Data Science and Engineering

Publication

Machine Learning

Volume

107

Issue

4

First Page

727

Last Page

747

ISSN

0885-6125

Identifier

10.1007/s10994-017-5676-y

Publisher

Springer Verlag (Germany)

Copyright Owner and License

Authors

Additional URL

https://doi.org/10.1007/s10994-017-5676-y

Download

Share

COinS