Practical Dataset Distillation Based on Deep Support Vectors

Practical Dataset Distillation Based on Deep Support Vectors

Conventional dataset distillation requires significant computational resources and assumes access to the entire dataset, an assumption impractical as it presumes all data resides on a central server. In this paper, we focus on dataset distillation in practical scenarios with access to only a fractio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2024-05
Hauptverfasser: Lee, Hyunho, Lee, Junhoo, Kwak, Nojun
Format: Artikel
Sprache:eng
Schlagworte:
Datasets
Distillation
Performance enhancement
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 Lee, Hyunho
Lee, Junhoo
Kwak, Nojun
description Conventional dataset distillation requires significant computational resources and assumes access to the entire dataset, an assumption impractical as it presumes all data resides on a central server. In this paper, we focus on dataset distillation in practical scenarios with access to only a fraction of the entire dataset. We introduce a novel distillation method that augments the conventional process by incorporating general model knowledge via the addition of Deep KKT (DKKT) loss. In practical settings, our approach showed improved performance compared to the baseline distribution matching distillation method on the CIFAR-10 dataset. Additionally, we present experimental evidence that Deep Support Vectors (DSVs) offer unique information to the original distillation, and their integration results in enhanced performance.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_3049908086</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3049908086</sourcerecordid><originalsourceid>FETCH-proquest_journals_30499080863</originalsourceid><addsrcrecordid>eNqNisEKAiEURSUIGmr-QWg9YDozOdCqLFoGRdtBzMBBRvM9_z8XfUCrezjnLkjFhdg1suV8RWqAiTHG-z3vOlGRwy1pg85oT5VGDRapcoDOe40uzPRY1IsWUNZGes8xhoT0aQ2GBBuyfGsPtv7tmmwv58fp2sQUPtkCjlPIaS5pFKwdBiaZ7MV_ry8iYjdl</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3049908086</pqid></control><display><type>article</type><title>Practical Dataset Distillation Based on Deep Support Vectors</title><source>Free E- Journals</source><creator>Lee, Hyunho ; Lee, Junhoo ; Kwak, Nojun</creator><creatorcontrib>Lee, Hyunho ; Lee, Junhoo ; Kwak, Nojun</creatorcontrib><description>Conventional dataset distillation requires significant computational resources and assumes access to the entire dataset, an assumption impractical as it presumes all data resides on a central server. In this paper, we focus on dataset distillation in practical scenarios with access to only a fraction of the entire dataset. We introduce a novel distillation method that augments the conventional process by incorporating general model knowledge via the addition of Deep KKT (DKKT) loss. In practical settings, our approach showed improved performance compared to the baseline distribution matching distillation method on the CIFAR-10 dataset. Additionally, we present experimental evidence that Deep Support Vectors (DSVs) offer unique information to the original distillation, and their integration results in enhanced performance.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Datasets ; Distillation ; Performance enhancement</subject><ispartof>arXiv.org, 2024-05</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (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>Lee, Hyunho</creatorcontrib><creatorcontrib>Lee, Junhoo</creatorcontrib><creatorcontrib>Kwak, Nojun</creatorcontrib><title>Practical Dataset Distillation Based on Deep Support Vectors</title><title>arXiv.org</title><description>Conventional dataset distillation requires significant computational resources and assumes access to the entire dataset, an assumption impractical as it presumes all data resides on a central server. In this paper, we focus on dataset distillation in practical scenarios with access to only a fraction of the entire dataset. We introduce a novel distillation method that augments the conventional process by incorporating general model knowledge via the addition of Deep KKT (DKKT) loss. In practical settings, our approach showed improved performance compared to the baseline distribution matching distillation method on the CIFAR-10 dataset. Additionally, we present experimental evidence that Deep Support Vectors (DSVs) offer unique information to the original distillation, and their integration results in enhanced performance.</description><subject>Datasets</subject><subject>Distillation</subject><subject>Performance enhancement</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNisEKAiEURSUIGmr-QWg9YDozOdCqLFoGRdtBzMBBRvM9_z8XfUCrezjnLkjFhdg1suV8RWqAiTHG-z3vOlGRwy1pg85oT5VGDRapcoDOe40uzPRY1IsWUNZGes8xhoT0aQ2GBBuyfGsPtv7tmmwv58fp2sQUPtkCjlPIaS5pFKwdBiaZ7MV_ry8iYjdl</recordid><startdate>20240501</startdate><enddate>20240501</enddate><creator>Lee, Hyunho</creator><creator>Lee, Junhoo</creator><creator>Kwak, Nojun</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>20240501</creationdate><title>Practical Dataset Distillation Based on Deep Support Vectors</title><author>Lee, Hyunho ; Lee, Junhoo ; Kwak, Nojun</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_30499080863</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Datasets</topic><topic>Distillation</topic><topic>Performance enhancement</topic><toplevel>online_resources</toplevel><creatorcontrib>Lee, Hyunho</creatorcontrib><creatorcontrib>Lee, Junhoo</creatorcontrib><creatorcontrib>Kwak, Nojun</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</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>Lee, Hyunho</au><au>Lee, Junhoo</au><au>Kwak, Nojun</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Practical Dataset Distillation Based on Deep Support Vectors</atitle><jtitle>arXiv.org</jtitle><date>2024-05-01</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Conventional dataset distillation requires significant computational resources and assumes access to the entire dataset, an assumption impractical as it presumes all data resides on a central server. In this paper, we focus on dataset distillation in practical scenarios with access to only a fraction of the entire dataset. We introduce a novel distillation method that augments the conventional process by incorporating general model knowledge via the addition of Deep KKT (DKKT) loss. In practical settings, our approach showed improved performance compared to the baseline distribution matching distillation method on the CIFAR-10 dataset. Additionally, we present experimental evidence that Deep Support Vectors (DSVs) offer unique information to the original distillation, and their integration results in enhanced performance.</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, 2024-05
issn 2331-8422
language eng
recordid cdi_proquest_journals_3049908086
source Free E- Journals
subjects Datasets
Distillation
Performance enhancement
title Practical Dataset Distillation Based on Deep Support Vectors
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-04T03%3A43%3A13IST&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=Practical%20Dataset%20Distillation%20Based%20on%20Deep%20Support%20Vectors&rft.jtitle=arXiv.org&rft.au=Lee,%20Hyunho&rft.date=2024-05-01&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E3049908086%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3049908086&rft_id=info:pmid/&rfr_iscdi=true