Accelerating Stochastic Gradient Descent Using Antithetic Sampling

(Mini-batch) Stochastic Gradient Descent is a popular optimization method which has been applied to many machine learning applications. But a rather high variance introduced by the stochastic gradient in each step may slow down the convergence. In this paper, we propose the antithetic sampling strat...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2018-10
Hauptverfasser: Liu, Jingchang, Xu, Linli
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Liu, Jingchang
Xu, Linli
description (Mini-batch) Stochastic Gradient Descent is a popular optimization method which has been applied to many machine learning applications. But a rather high variance introduced by the stochastic gradient in each step may slow down the convergence. In this paper, we propose the antithetic sampling strategy to reduce the variance by taking advantage of the internal structure in dataset. Under this new strategy, stochastic gradients in a mini-batch are no longer independent but negatively correlated as much as possible, while the mini-batch stochastic gradient is still an unbiased estimator of full gradient. For the binary classification problems, we just need to calculate the antithetic samples in advance, and reuse the result in each iteration, which is practical. Experiments are provided to confirm the effectiveness of the proposed method.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2117277914</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2117277914</sourcerecordid><originalsourceid>FETCH-proquest_journals_21172779143</originalsourceid><addsrcrecordid>eNqNyt0KgjAYxvERBEl5D0LHgtu01aF9n1vHMtZbTmyzva_3n0IX0NEfnuc3Y5GQkqfbXIgFixHbLMvERomikBHbl8ZAB0GTda-kIm8ajWRNcgn6YcFRcgQ0U-84idKRpQYmUel3343bis2fukOIf12y9fl0O1zTPvjPAEh164fgxqsWnCuh1I7n8j_1Bcy6OfI</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2117277914</pqid></control><display><type>article</type><title>Accelerating Stochastic Gradient Descent Using Antithetic Sampling</title><source>Free E- Journals</source><creator>Liu, Jingchang ; Xu, Linli</creator><creatorcontrib>Liu, Jingchang ; Xu, Linli</creatorcontrib><description>(Mini-batch) Stochastic Gradient Descent is a popular optimization method which has been applied to many machine learning applications. But a rather high variance introduced by the stochastic gradient in each step may slow down the convergence. In this paper, we propose the antithetic sampling strategy to reduce the variance by taking advantage of the internal structure in dataset. Under this new strategy, stochastic gradients in a mini-batch are no longer independent but negatively correlated as much as possible, while the mini-batch stochastic gradient is still an unbiased estimator of full gradient. For the binary classification problems, we just need to calculate the antithetic samples in advance, and reuse the result in each iteration, which is practical. Experiments are provided to confirm the effectiveness of the proposed method.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Machine learning ; Optimization ; Sampling</subject><ispartof>arXiv.org, 2018-10</ispartof><rights>2018. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Liu, Jingchang</creatorcontrib><creatorcontrib>Xu, Linli</creatorcontrib><title>Accelerating Stochastic Gradient Descent Using Antithetic Sampling</title><title>arXiv.org</title><description>(Mini-batch) Stochastic Gradient Descent is a popular optimization method which has been applied to many machine learning applications. But a rather high variance introduced by the stochastic gradient in each step may slow down the convergence. In this paper, we propose the antithetic sampling strategy to reduce the variance by taking advantage of the internal structure in dataset. Under this new strategy, stochastic gradients in a mini-batch are no longer independent but negatively correlated as much as possible, while the mini-batch stochastic gradient is still an unbiased estimator of full gradient. For the binary classification problems, we just need to calculate the antithetic samples in advance, and reuse the result in each iteration, which is practical. Experiments are provided to confirm the effectiveness of the proposed method.</description><subject>Machine learning</subject><subject>Optimization</subject><subject>Sampling</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNyt0KgjAYxvERBEl5D0LHgtu01aF9n1vHMtZbTmyzva_3n0IX0NEfnuc3Y5GQkqfbXIgFixHbLMvERomikBHbl8ZAB0GTda-kIm8ajWRNcgn6YcFRcgQ0U-84idKRpQYmUel3343bis2fukOIf12y9fl0O1zTPvjPAEh164fgxqsWnCuh1I7n8j_1Bcy6OfI</recordid><startdate>20181007</startdate><enddate>20181007</enddate><creator>Liu, Jingchang</creator><creator>Xu, Linli</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20181007</creationdate><title>Accelerating Stochastic Gradient Descent Using Antithetic Sampling</title><author>Liu, Jingchang ; Xu, Linli</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_21172779143</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Machine learning</topic><topic>Optimization</topic><topic>Sampling</topic><toplevel>online_resources</toplevel><creatorcontrib>Liu, Jingchang</creatorcontrib><creatorcontrib>Xu, Linli</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection (ProQuest)</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Liu, Jingchang</au><au>Xu, Linli</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Accelerating Stochastic Gradient Descent Using Antithetic Sampling</atitle><jtitle>arXiv.org</jtitle><date>2018-10-07</date><risdate>2018</risdate><eissn>2331-8422</eissn><abstract>(Mini-batch) Stochastic Gradient Descent is a popular optimization method which has been applied to many machine learning applications. But a rather high variance introduced by the stochastic gradient in each step may slow down the convergence. In this paper, we propose the antithetic sampling strategy to reduce the variance by taking advantage of the internal structure in dataset. Under this new strategy, stochastic gradients in a mini-batch are no longer independent but negatively correlated as much as possible, while the mini-batch stochastic gradient is still an unbiased estimator of full gradient. For the binary classification problems, we just need to calculate the antithetic samples in advance, and reuse the result in each iteration, which is practical. Experiments are provided to confirm the effectiveness of the proposed method.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2018-10
issn 2331-8422
language eng
recordid cdi_proquest_journals_2117277914
source Free E- Journals
subjects Machine learning
Optimization
Sampling
title Accelerating Stochastic Gradient Descent Using Antithetic Sampling
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-31T14%3A49%3A06IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Accelerating%20Stochastic%20Gradient%20Descent%20Using%20Antithetic%20Sampling&rft.jtitle=arXiv.org&rft.au=Liu,%20Jingchang&rft.date=2018-10-07&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2117277914%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2117277914&rft_id=info:pmid/&rfr_iscdi=true