Loading…
Using Large Ensembles of Control Variates for Variational Inference
Variational inference is increasingly being addressed with stochastic optimization. In this setting, the gradient's variance plays a crucial role in the optimization procedure, since high variance gradients lead to poor convergence. A popular approach used to reduce gradient's variance inv...
Saved in:
Published in: | arXiv.org 2020-10 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | |
---|---|
cites | |
container_end_page | |
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Geffner, Tomas Domke, Justin |
description | Variational inference is increasingly being addressed with stochastic optimization. In this setting, the gradient's variance plays a crucial role in the optimization procedure, since high variance gradients lead to poor convergence. A popular approach used to reduce gradient's variance involves the use of control variates. Despite the good results obtained, control variates developed for variational inference are typically looked at in isolation. In this paper we clarify the large number of control variates that are available by giving a systematic view of how they are derived. We also present a Bayesian risk minimization framework in which the quality of a procedure for combining control variates is quantified by its effect on optimization convergence rates, which leads to a very simple combination rule. Results show that combining a large number of control variates this way significantly improves the convergence of inference over using the typical gradient estimators or a reduced number of control variates. |
format | article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2127497667</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2127497667</sourcerecordid><originalsourceid>FETCH-proquest_journals_21274976673</originalsourceid><addsrcrecordid>eNqNi0EKwjAQRYMgWLR3CLgutJO20XWoKLhUtyXKpLTERGfa-9tFD-Dq8x7vr0QCShXZoQTYiJR5yPMcag1VpRJh7tyHTl4tdSibwPh-emQZnTQxjBS9fFjq7Tg7F2mBPgbr5SU4JAwv3Im1s54xXXYr9qfmZs7Zh-J3Qh7bIU40X7iFAnR51HWt1X_VD9RcOgI</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2127497667</pqid></control><display><type>article</type><title>Using Large Ensembles of Control Variates for Variational Inference</title><source>Publicly Available Content Database</source><creator>Geffner, Tomas ; Domke, Justin</creator><creatorcontrib>Geffner, Tomas ; Domke, Justin</creatorcontrib><description>Variational inference is increasingly being addressed with stochastic optimization. In this setting, the gradient's variance plays a crucial role in the optimization procedure, since high variance gradients lead to poor convergence. A popular approach used to reduce gradient's variance involves the use of control variates. Despite the good results obtained, control variates developed for variational inference are typically looked at in isolation. In this paper we clarify the large number of control variates that are available by giving a systematic view of how they are derived. We also present a Bayesian risk minimization framework in which the quality of a procedure for combining control variates is quantified by its effect on optimization convergence rates, which leads to a very simple combination rule. Results show that combining a large number of control variates this way significantly improves the convergence of inference over using the typical gradient estimators or a reduced number of control variates.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Bayesian analysis ; Convergence ; Inference ; Optimization</subject><ispartof>arXiv.org, 2020-10</ispartof><rights>2020. 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>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2127497667?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Geffner, Tomas</creatorcontrib><creatorcontrib>Domke, Justin</creatorcontrib><title>Using Large Ensembles of Control Variates for Variational Inference</title><title>arXiv.org</title><description>Variational inference is increasingly being addressed with stochastic optimization. In this setting, the gradient's variance plays a crucial role in the optimization procedure, since high variance gradients lead to poor convergence. A popular approach used to reduce gradient's variance involves the use of control variates. Despite the good results obtained, control variates developed for variational inference are typically looked at in isolation. In this paper we clarify the large number of control variates that are available by giving a systematic view of how they are derived. We also present a Bayesian risk minimization framework in which the quality of a procedure for combining control variates is quantified by its effect on optimization convergence rates, which leads to a very simple combination rule. Results show that combining a large number of control variates this way significantly improves the convergence of inference over using the typical gradient estimators or a reduced number of control variates.</description><subject>Bayesian analysis</subject><subject>Convergence</subject><subject>Inference</subject><subject>Optimization</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNi0EKwjAQRYMgWLR3CLgutJO20XWoKLhUtyXKpLTERGfa-9tFD-Dq8x7vr0QCShXZoQTYiJR5yPMcag1VpRJh7tyHTl4tdSibwPh-emQZnTQxjBS9fFjq7Tg7F2mBPgbr5SU4JAwv3Im1s54xXXYr9qfmZs7Zh-J3Qh7bIU40X7iFAnR51HWt1X_VD9RcOgI</recordid><startdate>20201022</startdate><enddate>20201022</enddate><creator>Geffner, Tomas</creator><creator>Domke, Justin</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>20201022</creationdate><title>Using Large Ensembles of Control Variates for Variational Inference</title><author>Geffner, Tomas ; Domke, Justin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_21274976673</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Bayesian analysis</topic><topic>Convergence</topic><topic>Inference</topic><topic>Optimization</topic><toplevel>online_resources</toplevel><creatorcontrib>Geffner, Tomas</creatorcontrib><creatorcontrib>Domke, Justin</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</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>Geffner, Tomas</au><au>Domke, Justin</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Using Large Ensembles of Control Variates for Variational Inference</atitle><jtitle>arXiv.org</jtitle><date>2020-10-22</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>Variational inference is increasingly being addressed with stochastic optimization. In this setting, the gradient's variance plays a crucial role in the optimization procedure, since high variance gradients lead to poor convergence. A popular approach used to reduce gradient's variance involves the use of control variates. Despite the good results obtained, control variates developed for variational inference are typically looked at in isolation. In this paper we clarify the large number of control variates that are available by giving a systematic view of how they are derived. We also present a Bayesian risk minimization framework in which the quality of a procedure for combining control variates is quantified by its effect on optimization convergence rates, which leads to a very simple combination rule. Results show that combining a large number of control variates this way significantly improves the convergence of inference over using the typical gradient estimators or a reduced number of control variates.</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, 2020-10 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2127497667 |
source | Publicly Available Content Database |
subjects | Bayesian analysis Convergence Inference Optimization |
title | Using Large Ensembles of Control Variates for Variational Inference |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-26T15%3A54%3A32IST&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=Using%20Large%20Ensembles%20of%20Control%20Variates%20for%20Variational%20Inference&rft.jtitle=arXiv.org&rft.au=Geffner,%20Tomas&rft.date=2020-10-22&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2127497667%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_21274976673%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2127497667&rft_id=info:pmid/&rfr_iscdi=true |