Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction

In this paper, we study the lower complexity bounds for finite-sum optimization problems, where the objective is the average of $n$ individual component functions. We consider Proximal Incremental First-order (PIFO) algorithms which have access to the gradient and proximal oracles for each component...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Han, Yuze, Xie, Guangzeng, Zhang, Zhihua
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Han, Yuze
Xie, Guangzeng
Zhang, Zhihua
description In this paper, we study the lower complexity bounds for finite-sum optimization problems, where the objective is the average of $n$ individual component functions. We consider Proximal Incremental First-order (PIFO) algorithms which have access to the gradient and proximal oracles for each component function. To incorporate loopless methods, we also allow PIFO algorithms to obtain the full gradient infrequently. We develop a novel approach to constructing the hard instances, which partitions the tridiagonal matrix of classical examples into $n$ groups. This construction is friendly to the analysis of PIFO algorithms. Based on this construction, we establish the lower complexity bounds for finite-sum minimax optimization problems when the objective is convex-concave or nonconvex-strongly-concave and the class of component functions is $L$-average smooth. Most of these bounds are nearly matched by existing upper bounds up to log factors. We can also derive similar lower bounds for finite-sum minimization problems as previous work under both smoothness and average smoothness assumptions. Our lower bounds imply that proximal oracles for smooth functions are not much more powerful than gradient oracles.
doi_str_mv 10.48550/arxiv.2103.08280
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2103_08280</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2103_08280</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-8698c741ca8f005891f45e99563468132c4589997d2af726f743161a298517793</originalsourceid><addsrcrecordid>eNotj8tKxDAARbNxIaMf4GryA615P9xpcVQojDjduCqxTTDQNiVJdcavd2Z0deFwOXAAuMGoZIpzdGvi3n-VBCNaIkUUugTvdfi2EVZhnAe79_kAH8Iy9QkGBzd-8tkWu2WE2zn70f-Y7MMEX2P4GOyY7mDzaeGbTcuQEzRTf9RMKcelO92uwIUzQ7LX_7sCzeaxqZ6Levv0Ut3XhRESFUpo1UmGO6McQlxp7Bi3WnNBmVCYko4dodayJ8ZJIpxkFAtsiFYcS6npCqz_tOe2do5-NPHQnhrbcyP9BQtZSxo</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction</title><source>arXiv.org</source><creator>Han, Yuze ; Xie, Guangzeng ; Zhang, Zhihua</creator><creatorcontrib>Han, Yuze ; Xie, Guangzeng ; Zhang, Zhihua</creatorcontrib><description>In this paper, we study the lower complexity bounds for finite-sum optimization problems, where the objective is the average of $n$ individual component functions. We consider Proximal Incremental First-order (PIFO) algorithms which have access to the gradient and proximal oracles for each component function. To incorporate loopless methods, we also allow PIFO algorithms to obtain the full gradient infrequently. We develop a novel approach to constructing the hard instances, which partitions the tridiagonal matrix of classical examples into $n$ groups. This construction is friendly to the analysis of PIFO algorithms. Based on this construction, we establish the lower complexity bounds for finite-sum minimax optimization problems when the objective is convex-concave or nonconvex-strongly-concave and the class of component functions is $L$-average smooth. Most of these bounds are nearly matched by existing upper bounds up to log factors. We can also derive similar lower bounds for finite-sum minimization problems as previous work under both smoothness and average smoothness assumptions. Our lower bounds imply that proximal oracles for smooth functions are not much more powerful than gradient oracles.</description><identifier>DOI: 10.48550/arxiv.2103.08280</identifier><language>eng</language><subject>Computer Science - Learning ; Mathematics - Optimization and Control ; Statistics - Machine Learning</subject><creationdate>2021-03</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2103.08280$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2103.08280$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Han, Yuze</creatorcontrib><creatorcontrib>Xie, Guangzeng</creatorcontrib><creatorcontrib>Zhang, Zhihua</creatorcontrib><title>Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction</title><description>In this paper, we study the lower complexity bounds for finite-sum optimization problems, where the objective is the average of $n$ individual component functions. We consider Proximal Incremental First-order (PIFO) algorithms which have access to the gradient and proximal oracles for each component function. To incorporate loopless methods, we also allow PIFO algorithms to obtain the full gradient infrequently. We develop a novel approach to constructing the hard instances, which partitions the tridiagonal matrix of classical examples into $n$ groups. This construction is friendly to the analysis of PIFO algorithms. Based on this construction, we establish the lower complexity bounds for finite-sum minimax optimization problems when the objective is convex-concave or nonconvex-strongly-concave and the class of component functions is $L$-average smooth. Most of these bounds are nearly matched by existing upper bounds up to log factors. We can also derive similar lower bounds for finite-sum minimization problems as previous work under both smoothness and average smoothness assumptions. Our lower bounds imply that proximal oracles for smooth functions are not much more powerful than gradient oracles.</description><subject>Computer Science - Learning</subject><subject>Mathematics - Optimization and Control</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj8tKxDAARbNxIaMf4GryA615P9xpcVQojDjduCqxTTDQNiVJdcavd2Z0deFwOXAAuMGoZIpzdGvi3n-VBCNaIkUUugTvdfi2EVZhnAe79_kAH8Iy9QkGBzd-8tkWu2WE2zn70f-Y7MMEX2P4GOyY7mDzaeGbTcuQEzRTf9RMKcelO92uwIUzQ7LX_7sCzeaxqZ6Levv0Ut3XhRESFUpo1UmGO6McQlxp7Bi3WnNBmVCYko4dodayJ8ZJIpxkFAtsiFYcS6npCqz_tOe2do5-NPHQnhrbcyP9BQtZSxo</recordid><startdate>20210315</startdate><enddate>20210315</enddate><creator>Han, Yuze</creator><creator>Xie, Guangzeng</creator><creator>Zhang, Zhihua</creator><scope>AKY</scope><scope>AKZ</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20210315</creationdate><title>Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction</title><author>Han, Yuze ; Xie, Guangzeng ; Zhang, Zhihua</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-8698c741ca8f005891f45e99563468132c4589997d2af726f743161a298517793</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Computer Science - Learning</topic><topic>Mathematics - Optimization and Control</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Han, Yuze</creatorcontrib><creatorcontrib>Xie, Guangzeng</creatorcontrib><creatorcontrib>Zhang, Zhihua</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Mathematics</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Han, Yuze</au><au>Xie, Guangzeng</au><au>Zhang, Zhihua</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction</atitle><date>2021-03-15</date><risdate>2021</risdate><abstract>In this paper, we study the lower complexity bounds for finite-sum optimization problems, where the objective is the average of $n$ individual component functions. We consider Proximal Incremental First-order (PIFO) algorithms which have access to the gradient and proximal oracles for each component function. To incorporate loopless methods, we also allow PIFO algorithms to obtain the full gradient infrequently. We develop a novel approach to constructing the hard instances, which partitions the tridiagonal matrix of classical examples into $n$ groups. This construction is friendly to the analysis of PIFO algorithms. Based on this construction, we establish the lower complexity bounds for finite-sum minimax optimization problems when the objective is convex-concave or nonconvex-strongly-concave and the class of component functions is $L$-average smooth. Most of these bounds are nearly matched by existing upper bounds up to log factors. We can also derive similar lower bounds for finite-sum minimization problems as previous work under both smoothness and average smoothness assumptions. Our lower bounds imply that proximal oracles for smooth functions are not much more powerful than gradient oracles.</abstract><doi>10.48550/arxiv.2103.08280</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2103.08280
ispartof
issn
language eng
recordid cdi_arxiv_primary_2103_08280
source arXiv.org
subjects Computer Science - Learning
Mathematics - Optimization and Control
Statistics - Machine Learning
title Lower Complexity Bounds of Finite-Sum Optimization Problems: The Results and Construction
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-26T19%3A09%3A41IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Lower%20Complexity%20Bounds%20of%20Finite-Sum%20Optimization%20Problems:%20The%20Results%20and%20Construction&rft.au=Han,%20Yuze&rft.date=2021-03-15&rft_id=info:doi/10.48550/arxiv.2103.08280&rft_dat=%3Carxiv_GOX%3E2103_08280%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true