Parallel Bayesian Active Learning using Dropout for Optimizing High-Speed Channel Equalization

This work realizes the parallelization of Bayesian Active Learning using Dropout (BAL-DO) and is successfully applied for optimizing equalization settings for high-speed channel receivers (RX). In this paper, parallel BAL-DO can achieve the largest horizontal eye (HEYE) opening and its corresponding...

Full description

Saved in:
Bibliographic Details
Main Authors: Yang, Xianbo, Torun, Hakki M., Tang, Junyan, Paladhi, Pavel Roy, Zhang, Yanyan, Becker, Wiren D., Hejase, Jose A., Swaminathan, Madhavan
Format: Conference Proceeding
Language:English
Subjects:
Bayes methods
Conferences
Electric potential
Optical signal processing
Optimization
Receivers
Time-domain analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 3
container_issue
container_start_page 1
container_title
container_volume
creator Yang, Xianbo
Torun, Hakki M.
Tang, Junyan
Paladhi, Pavel Roy
Zhang, Yanyan
Becker, Wiren D.
Hejase, Jose A.
Swaminathan, Madhavan
description This work realizes the parallelization of Bayesian Active Learning using Dropout (BAL-DO) and is successfully applied for optimizing equalization settings for high-speed channel receivers (RX). In this paper, parallel BAL-DO can achieve the largest horizontal eye (HEYE) opening and its corresponding equalization setting 12 times faster, on average, than the previously reported sequential BAL-DO. This corresponds to an average of 40-50 times faster than standard exhaustive time-domain simulations. Moreover, the HEYE prediction accuracy across the whole design space is close to 3 times better while using parallelization than sequential BAL-DO. With these outstanding results, parallel BAL-DO dramatically improves the efficiency for RX equalization optimization and greatly reduces the computing resources and time needed.
doi_str_mv 10.1109/EPEPS51341.2021.9609205
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_9609205</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9609205</ieee_id><sourcerecordid>9609205</sourcerecordid><originalsourceid>FETCH-LOGICAL-i203t-5342075f21002543ae72147c2d86900f18c03368ad49489f17a3fbe1cd81de4d3</originalsourceid><addsrcrecordid>eNotkM1OwkAURkcTEwn2CVw4L1C8d_46s8RawYQEEnQrGdtbGFPa2hYTeHolsvlOchZn8TH2gDBBBPeYrbLVWqNUOBEgcOIMOAH6ikUusWiMVkoYp6_ZSKDRsULUtyzq-y8AkGitdWbEPla-81VFFX_yR-qDr_k0H8IP8QX5rg71lh_68z53TdscBl42HV-2Q9iH01nPw3YXr1uigqc7X9d_oez74Ktw8kNo6jt2U_qqp-jCMXt_yd7SebxYzl7T6SIOAuQQa6kEJLoUCCC0kp4SgSrJRWGNAyjR5iClsb5QTllXYuJl-UmYFxYLUoUcs_v_biCiTduFve-Om8sj8heRiVWF</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Parallel Bayesian Active Learning using Dropout for Optimizing High-Speed Channel Equalization</title><source>IEEE Xplore All Conference Series</source><creator>Yang, Xianbo ; Torun, Hakki M. ; Tang, Junyan ; Paladhi, Pavel Roy ; Zhang, Yanyan ; Becker, Wiren D. ; Hejase, Jose A. ; Swaminathan, Madhavan</creator><creatorcontrib>Yang, Xianbo ; Torun, Hakki M. ; Tang, Junyan ; Paladhi, Pavel Roy ; Zhang, Yanyan ; Becker, Wiren D. ; Hejase, Jose A. ; Swaminathan, Madhavan</creatorcontrib><description>This work realizes the parallelization of Bayesian Active Learning using Dropout (BAL-DO) and is successfully applied for optimizing equalization settings for high-speed channel receivers (RX). In this paper, parallel BAL-DO can achieve the largest horizontal eye (HEYE) opening and its corresponding equalization setting 12 times faster, on average, than the previously reported sequential BAL-DO. This corresponds to an average of 40-50 times faster than standard exhaustive time-domain simulations. Moreover, the HEYE prediction accuracy across the whole design space is close to 3 times better while using parallelization than sequential BAL-DO. With these outstanding results, parallel BAL-DO dramatically improves the efficiency for RX equalization optimization and greatly reduces the computing resources and time needed.</description><identifier>EISSN: 2165-4115</identifier><identifier>EISBN: 9781665442695</identifier><identifier>EISBN: 1665442697</identifier><identifier>DOI: 10.1109/EPEPS51341.2021.9609205</identifier><language>eng</language><publisher>IEEE</publisher><subject>Bayes methods ; Conferences ; Electric potential ; Optical signal processing ; Optimization ; Receivers ; Time-domain analysis</subject><ispartof>2021 IEEE 30th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS), 2021, p.1-3</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9609205$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,27924,54554,54931</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9609205$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Yang, Xianbo</creatorcontrib><creatorcontrib>Torun, Hakki M.</creatorcontrib><creatorcontrib>Tang, Junyan</creatorcontrib><creatorcontrib>Paladhi, Pavel Roy</creatorcontrib><creatorcontrib>Zhang, Yanyan</creatorcontrib><creatorcontrib>Becker, Wiren D.</creatorcontrib><creatorcontrib>Hejase, Jose A.</creatorcontrib><creatorcontrib>Swaminathan, Madhavan</creatorcontrib><title>Parallel Bayesian Active Learning using Dropout for Optimizing High-Speed Channel Equalization</title><title>2021 IEEE 30th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)</title><addtitle>EPEPS</addtitle><description>This work realizes the parallelization of Bayesian Active Learning using Dropout (BAL-DO) and is successfully applied for optimizing equalization settings for high-speed channel receivers (RX). In this paper, parallel BAL-DO can achieve the largest horizontal eye (HEYE) opening and its corresponding equalization setting 12 times faster, on average, than the previously reported sequential BAL-DO. This corresponds to an average of 40-50 times faster than standard exhaustive time-domain simulations. Moreover, the HEYE prediction accuracy across the whole design space is close to 3 times better while using parallelization than sequential BAL-DO. With these outstanding results, parallel BAL-DO dramatically improves the efficiency for RX equalization optimization and greatly reduces the computing resources and time needed.</description><subject>Bayes methods</subject><subject>Conferences</subject><subject>Electric potential</subject><subject>Optical signal processing</subject><subject>Optimization</subject><subject>Receivers</subject><subject>Time-domain analysis</subject><issn>2165-4115</issn><isbn>9781665442695</isbn><isbn>1665442697</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2021</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotkM1OwkAURkcTEwn2CVw4L1C8d_46s8RawYQEEnQrGdtbGFPa2hYTeHolsvlOchZn8TH2gDBBBPeYrbLVWqNUOBEgcOIMOAH6ikUusWiMVkoYp6_ZSKDRsULUtyzq-y8AkGitdWbEPla-81VFFX_yR-qDr_k0H8IP8QX5rg71lh_68z53TdscBl42HV-2Q9iH01nPw3YXr1uigqc7X9d_oez74Ktw8kNo6jt2U_qqp-jCMXt_yd7SebxYzl7T6SIOAuQQa6kEJLoUCCC0kp4SgSrJRWGNAyjR5iClsb5QTllXYuJl-UmYFxYLUoUcs_v_biCiTduFve-Om8sj8heRiVWF</recordid><startdate>20211017</startdate><enddate>20211017</enddate><creator>Yang, Xianbo</creator><creator>Torun, Hakki M.</creator><creator>Tang, Junyan</creator><creator>Paladhi, Pavel Roy</creator><creator>Zhang, Yanyan</creator><creator>Becker, Wiren D.</creator><creator>Hejase, Jose A.</creator><creator>Swaminathan, Madhavan</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>20211017</creationdate><title>Parallel Bayesian Active Learning using Dropout for Optimizing High-Speed Channel Equalization</title><author>Yang, Xianbo ; Torun, Hakki M. ; Tang, Junyan ; Paladhi, Pavel Roy ; Zhang, Yanyan ; Becker, Wiren D. ; Hejase, Jose A. ; Swaminathan, Madhavan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i203t-5342075f21002543ae72147c2d86900f18c03368ad49489f17a3fbe1cd81de4d3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Bayes methods</topic><topic>Conferences</topic><topic>Electric potential</topic><topic>Optical signal processing</topic><topic>Optimization</topic><topic>Receivers</topic><topic>Time-domain analysis</topic><toplevel>online_resources</toplevel><creatorcontrib>Yang, Xianbo</creatorcontrib><creatorcontrib>Torun, Hakki M.</creatorcontrib><creatorcontrib>Tang, Junyan</creatorcontrib><creatorcontrib>Paladhi, Pavel Roy</creatorcontrib><creatorcontrib>Zhang, Yanyan</creatorcontrib><creatorcontrib>Becker, Wiren D.</creatorcontrib><creatorcontrib>Hejase, Jose A.</creatorcontrib><creatorcontrib>Swaminathan, Madhavan</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>Yang, Xianbo</au><au>Torun, Hakki M.</au><au>Tang, Junyan</au><au>Paladhi, Pavel Roy</au><au>Zhang, Yanyan</au><au>Becker, Wiren D.</au><au>Hejase, Jose A.</au><au>Swaminathan, Madhavan</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Parallel Bayesian Active Learning using Dropout for Optimizing High-Speed Channel Equalization</atitle><btitle>2021 IEEE 30th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)</btitle><stitle>EPEPS</stitle><date>2021-10-17</date><risdate>2021</risdate><spage>1</spage><epage>3</epage><pages>1-3</pages><eissn>2165-4115</eissn><eisbn>9781665442695</eisbn><eisbn>1665442697</eisbn><abstract>This work realizes the parallelization of Bayesian Active Learning using Dropout (BAL-DO) and is successfully applied for optimizing equalization settings for high-speed channel receivers (RX). In this paper, parallel BAL-DO can achieve the largest horizontal eye (HEYE) opening and its corresponding equalization setting 12 times faster, on average, than the previously reported sequential BAL-DO. This corresponds to an average of 40-50 times faster than standard exhaustive time-domain simulations. Moreover, the HEYE prediction accuracy across the whole design space is close to 3 times better while using parallelization than sequential BAL-DO. With these outstanding results, parallel BAL-DO dramatically improves the efficiency for RX equalization optimization and greatly reduces the computing resources and time needed.</abstract><pub>IEEE</pub><doi>10.1109/EPEPS51341.2021.9609205</doi><tpages>3</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2165-4115
ispartof 2021 IEEE 30th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS), 2021, p.1-3
issn 2165-4115
language eng
recordid cdi_ieee_primary_9609205
source IEEE Xplore All Conference Series
subjects Bayes methods
Conferences
Electric potential
Optical signal processing
Optimization
Receivers
Time-domain analysis
title Parallel Bayesian Active Learning using Dropout for Optimizing High-Speed Channel Equalization
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-12T21%3A10%3A04IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Parallel%20Bayesian%20Active%20Learning%20using%20Dropout%20for%20Optimizing%20High-Speed%20Channel%20Equalization&rft.btitle=2021%20IEEE%2030th%20Conference%20on%20Electrical%20Performance%20of%20Electronic%20Packaging%20and%20Systems%20(EPEPS)&rft.au=Yang,%20Xianbo&rft.date=2021-10-17&rft.spage=1&rft.epage=3&rft.pages=1-3&rft.eissn=2165-4115&rft_id=info:doi/10.1109/EPEPS51341.2021.9609205&rft.eisbn=9781665442695&rft.eisbn_list=1665442697&rft_dat=%3Cieee_CHZPO%3E9609205%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i203t-5342075f21002543ae72147c2d86900f18c03368ad49489f17a3fbe1cd81de4d3%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=9609205&rfr_iscdi=true