Please use this identifier to cite or link to this item:
|Title:||Distributed multi-task classification : a decentralized online learning approach||Authors:||Zhang, Chi
Soh, Yeng Chai
Lee, Bu Sung
Hoi, Steven C. H.
|Keywords:||Decentralized Distributed Learning
Engineering::Electrical and electronic engineering
|Issue Date:||2018||Source:||Zhang, C., Zhao, P., Hao, S., Soh, Y. C., Lee, B. S., Miao, C., & Hoi, S. C. H. (2018). Distributed multi-task classification: a decentralized online learning approach. Machine Learning, 107(4), 727-747. doi:10.1007/s10994-017-5676-y||Series/Report no.:||Machine Learning||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 O(T−−√) regret bound when compared with the best classifier in hindsight, which is further validated by experiments on both synthetic and real-world datasets.||URI:||https://hdl.handle.net/10356/82962
|ISSN:||0885-6125||DOI:||http://dx.doi.org/10.1007/s10994-017-5676-y||Rights:||© 2017 The Author(s). All rights reserved.||metadata.item.grantfulltext:||none||metadata.item.fulltext:||No Fulltext|
|Appears in Collections:||SCSE Journal Articles|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.