Multi-task learning on the edge: cost-efficiency and theoretical optimality

This article proposes a distributed multi-task learning (MTL) algorithm based on supervised principal component analysis (SPCA) which is: (i) theoretically optimal for Gaussian mixtures, (ii) computationally cheap and scalable. Supporting experiments on synthetic and real benchmark data demonstrate...

Full description

Bibliographic Details
Main Authors: Fakhry, Sami, Couillet, Romain, Tiomoko, Malik
Format: Article in Journal/Newspaper
Language:unknown
Published: arXiv 2021
Subjects:
Machine Learning cs.LG
FOS Computer and information sciences
sami
Online Access:https://dx.doi.org/10.48550/arxiv.2110.04639
https://arxiv.org/abs/2110.04639
id ftdatacite:10.48550/arxiv.2110.04639
record_format openpolar
spelling ftdatacite:10.48550/arxiv.2110.04639 2023-05-15T18:11:31+02:00 Multi-task learning on the edge: cost-efficiency and theoretical optimality Fakhry, Sami Couillet, Romain Tiomoko, Malik 2021 https://dx.doi.org/10.48550/arxiv.2110.04639 https://arxiv.org/abs/2110.04639 unknown arXiv arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ Machine Learning cs.LG FOS Computer and information sciences Article CreativeWork article Preprint 2021 ftdatacite https://doi.org/10.48550/arxiv.2110.04639 2022-03-10T13:46:11Z This article proposes a distributed multi-task learning (MTL) algorithm based on supervised principal component analysis (SPCA) which is: (i) theoretically optimal for Gaussian mixtures, (ii) computationally cheap and scalable. Supporting experiments on synthetic and real benchmark data demonstrate that significant energy gains can be obtained with no performance loss. : 4 pages, 5 figures, code to reproduce figure available at: https://github.com/Sami-fak/DistributedMTLSPCA Article in Journal/Newspaper sami DataCite Metadata Store (German National Library of Science and Technology)
institution Open Polar
collection DataCite Metadata Store (German National Library of Science and Technology)
op_collection_id ftdatacite
language unknown
topic Machine Learning cs.LG
FOS Computer and information sciences
spellingShingle Machine Learning cs.LG
FOS Computer and information sciences
Fakhry, Sami
Couillet, Romain
Tiomoko, Malik
Multi-task learning on the edge: cost-efficiency and theoretical optimality
topic_facet Machine Learning cs.LG
FOS Computer and information sciences
description This article proposes a distributed multi-task learning (MTL) algorithm based on supervised principal component analysis (SPCA) which is: (i) theoretically optimal for Gaussian mixtures, (ii) computationally cheap and scalable. Supporting experiments on synthetic and real benchmark data demonstrate that significant energy gains can be obtained with no performance loss. : 4 pages, 5 figures, code to reproduce figure available at: https://github.com/Sami-fak/DistributedMTLSPCA
format Article in Journal/Newspaper
author Fakhry, Sami
Couillet, Romain
Tiomoko, Malik
author_facet Fakhry, Sami
Couillet, Romain
Tiomoko, Malik
author_sort Fakhry, Sami
title Multi-task learning on the edge: cost-efficiency and theoretical optimality
title_short Multi-task learning on the edge: cost-efficiency and theoretical optimality
title_full Multi-task learning on the edge: cost-efficiency and theoretical optimality
title_fullStr Multi-task learning on the edge: cost-efficiency and theoretical optimality
title_full_unstemmed Multi-task learning on the edge: cost-efficiency and theoretical optimality
title_sort multi-task learning on the edge: cost-efficiency and theoretical optimality
publisher arXiv
publishDate 2021
url https://dx.doi.org/10.48550/arxiv.2110.04639
https://arxiv.org/abs/2110.04639
genre sami
genre_facet sami
op_rights arXiv.org perpetual, non-exclusive license
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
op_doi https://doi.org/10.48550/arxiv.2110.04639
_version_ 1766184167280738304

Similar Items