Diffusion LMS for multitask problems with overlapping hypothesis subspaces

There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperat...

Full description

Saved in:
Bibliographic Details
Main Authors: Jie Chen, Richard, Cedric, Hero, Alfred O., Sayed, Ali H.
Format: Conference Proceeding
Language:English
Subjects:
Adaptive systems
collaborative processing
Convergence
diffusion strategy
distributed optimization
Estimation
Least squares approximations
Multitask learning
Optimization
Signal processing algorithms
Vectors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 6
container_issue
container_start_page 1
container_title
container_volume
creator Jie Chen
Richard, Cedric
Hero, Alfred O.
Sayed, Ali H.
description There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperative algorithm based on diffusion adaptation is derived. Some results on its stability and convergence properties are also provided. Simulations are conducted to illustrate the theoretical results.
doi_str_mv 10.1109/MLSP.2014.6958929
format conference_proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_6958929</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6958929</ieee_id><sourcerecordid>6958929</sourcerecordid><originalsourceid>FETCH-LOGICAL-i218t-c449b2455063464f157ea25eecd1a8b810d5e0d7fdd52ef9ce94ac088483abaf3</originalsourceid><addsrcrecordid>eNot0L1OwzAUQGGDQKIqfQDE4hdI8fVP7DuiAgWUCqSCxFY5yTU1JE0Up6C-PQOdzvYNh7ErEHMAgTerYv06lwL0PEfjUOIJm6F1oC2iylHnp2wilXUZSvdxxiZgDGTSaLhgs5S-hBCAubVOTdjzXQxhn2K348VqzUM38HbfjHH06Zv3Q1c21Cb-G8ct735oaHzfx90n3x76btxSiomnfZl6X1G6ZOfBN4lmx07Z-8P92-IxK16WT4vbIosS3JhVWmMptTEiVzrXAYwlLw1RVYN3pQNRGxK1DXVtJAWsCLWvhHPaKV_6oKbs-t-NRLTph9j64bA5nlB_Gf5RlQ</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Diffusion LMS for multitask problems with overlapping hypothesis subspaces</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Jie Chen ; Richard, Cedric ; Hero, Alfred O. ; Sayed, Ali H.</creator><creatorcontrib>Jie Chen ; Richard, Cedric ; Hero, Alfred O. ; Sayed, Ali H.</creatorcontrib><description>There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperative algorithm based on diffusion adaptation is derived. Some results on its stability and convergence properties are also provided. Simulations are conducted to illustrate the theoretical results.</description><identifier>ISSN: 1551-2541</identifier><identifier>EISSN: 2378-928X</identifier><identifier>EISBN: 9781479936946</identifier><identifier>EISBN: 1479936944</identifier><identifier>DOI: 10.1109/MLSP.2014.6958929</identifier><language>eng</language><publisher>IEEE</publisher><subject>Adaptive systems ; collaborative processing ; Convergence ; diffusion strategy ; distributed optimization ; Estimation ; Least squares approximations ; Multitask learning ; Optimization ; Signal processing algorithms ; Vectors</subject><ispartof>2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 2014, p.1-6</ispartof><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6958929$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54555,54920,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6958929$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Jie Chen</creatorcontrib><creatorcontrib>Richard, Cedric</creatorcontrib><creatorcontrib>Hero, Alfred O.</creatorcontrib><creatorcontrib>Sayed, Ali H.</creatorcontrib><title>Diffusion LMS for multitask problems with overlapping hypothesis subspaces</title><title>2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP)</title><addtitle>MLSP</addtitle><description>There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperative algorithm based on diffusion adaptation is derived. Some results on its stability and convergence properties are also provided. Simulations are conducted to illustrate the theoretical results.</description><subject>Adaptive systems</subject><subject>collaborative processing</subject><subject>Convergence</subject><subject>diffusion strategy</subject><subject>distributed optimization</subject><subject>Estimation</subject><subject>Least squares approximations</subject><subject>Multitask learning</subject><subject>Optimization</subject><subject>Signal processing algorithms</subject><subject>Vectors</subject><issn>1551-2541</issn><issn>2378-928X</issn><isbn>9781479936946</isbn><isbn>1479936944</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2014</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNot0L1OwzAUQGGDQKIqfQDE4hdI8fVP7DuiAgWUCqSCxFY5yTU1JE0Up6C-PQOdzvYNh7ErEHMAgTerYv06lwL0PEfjUOIJm6F1oC2iylHnp2wilXUZSvdxxiZgDGTSaLhgs5S-hBCAubVOTdjzXQxhn2K348VqzUM38HbfjHH06Zv3Q1c21Cb-G8ct735oaHzfx90n3x76btxSiomnfZl6X1G6ZOfBN4lmx07Z-8P92-IxK16WT4vbIosS3JhVWmMptTEiVzrXAYwlLw1RVYN3pQNRGxK1DXVtJAWsCLWvhHPaKV_6oKbs-t-NRLTph9j64bA5nlB_Gf5RlQ</recordid><startdate>20141114</startdate><enddate>20141114</enddate><creator>Jie Chen</creator><creator>Richard, Cedric</creator><creator>Hero, Alfred O.</creator><creator>Sayed, Ali H.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>20141114</creationdate><title>Diffusion LMS for multitask problems with overlapping hypothesis subspaces</title><author>Jie Chen ; Richard, Cedric ; Hero, Alfred O. ; Sayed, Ali H.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i218t-c449b2455063464f157ea25eecd1a8b810d5e0d7fdd52ef9ce94ac088483abaf3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2014</creationdate><topic>Adaptive systems</topic><topic>collaborative processing</topic><topic>Convergence</topic><topic>diffusion strategy</topic><topic>distributed optimization</topic><topic>Estimation</topic><topic>Least squares approximations</topic><topic>Multitask learning</topic><topic>Optimization</topic><topic>Signal processing algorithms</topic><topic>Vectors</topic><toplevel>online_resources</toplevel><creatorcontrib>Jie Chen</creatorcontrib><creatorcontrib>Richard, Cedric</creatorcontrib><creatorcontrib>Hero, Alfred O.</creatorcontrib><creatorcontrib>Sayed, Ali H.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Jie Chen</au><au>Richard, Cedric</au><au>Hero, Alfred O.</au><au>Sayed, Ali H.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Diffusion LMS for multitask problems with overlapping hypothesis subspaces</atitle><btitle>2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP)</btitle><stitle>MLSP</stitle><date>2014-11-14</date><risdate>2014</risdate><spage>1</spage><epage>6</epage><pages>1-6</pages><issn>1551-2541</issn><eissn>2378-928X</eissn><eisbn>9781479936946</eisbn><eisbn>1479936944</eisbn><abstract>There are many important applications that are multitask-oriented in the sense that there are multiple optimum parameter vectors to be inferred simultaneously by networked agents. In this paper, we formulate an online multitask learning problem where node hypothesis spaces partly overlap. A cooperative algorithm based on diffusion adaptation is derived. Some results on its stability and convergence properties are also provided. Simulations are conducted to illustrate the theoretical results.</abstract><pub>IEEE</pub><doi>10.1109/MLSP.2014.6958929</doi><tpages>6</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 1551-2541
ispartof 2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 2014, p.1-6
issn 1551-2541
2378-928X
language eng
recordid cdi_ieee_primary_6958929
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Adaptive systems
collaborative processing
Convergence
diffusion strategy
distributed optimization
Estimation
Least squares approximations
Multitask learning
Optimization
Signal processing algorithms
Vectors
title Diffusion LMS for multitask problems with overlapping hypothesis subspaces
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T21%3A10%3A35IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Diffusion%20LMS%20for%20multitask%20problems%20with%20overlapping%20hypothesis%20subspaces&rft.btitle=2014%20IEEE%20International%20Workshop%20on%20Machine%20Learning%20for%20Signal%20Processing%20(MLSP)&rft.au=Jie%20Chen&rft.date=2014-11-14&rft.spage=1&rft.epage=6&rft.pages=1-6&rft.issn=1551-2541&rft.eissn=2378-928X&rft_id=info:doi/10.1109/MLSP.2014.6958929&rft.eisbn=9781479936946&rft.eisbn_list=1479936944&rft_dat=%3Cieee_6IE%3E6958929%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i218t-c449b2455063464f157ea25eecd1a8b810d5e0d7fdd52ef9ce94ac088483abaf3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=6958929&rfr_iscdi=true