Towards Task and Architecture-Independent Generalization Gap Predictors

Towards Task and Architecture-Independent Generalization Gap Predictors

Can we use deep learning to predict when deep learning works? Our results suggest the affirmative. We created a dataset by training 13,500 neural networks with different architectures, on different variations of spiral datasets, and using different optimization parameters. We used this dataset to tr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2019-06
Hauptverfasser: Yak, Scott, Gonzalvo, Javier, Mazzawi, Hanna
Format: Artikel
Sprache:eng
Schlagworte:
Architecture
Machine learning
Neural networks
Optimization
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 Yak, Scott
Gonzalvo, Javier
Mazzawi, Hanna
description Can we use deep learning to predict when deep learning works? Our results suggest the affirmative. We created a dataset by training 13,500 neural networks with different architectures, on different variations of spiral datasets, and using different optimization parameters. We used this dataset to train task-independent and architecture-independent generalization gap predictors for those neural networks. We extend Jiang et al. (2018) to also use DNNs and RNNs and show that they outperform the linear model, obtaining \(R^2=0.965\). We also show results for architecture-independent, task-independent, and out-of-distribution generalization gap prediction tasks. Both DNNs and RNNs consistently and significantly outperform linear models, with RNNs obtaining \(R^2=0.584\).
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2235435523</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2235435523</sourcerecordid><originalsourceid>FETCH-proquest_journals_22354355233</originalsourceid><addsrcrecordid>eNqNjrEKwjAURYMgWLT_EHAu1KRRVxGtbg7dyyN5YmpJ6nspgl9vBz_A5Z7hnOHORKa03hT7SqmFyJm7sizVdqeM0Zmom_gGciwb4KeE4OSB7MMntGkkLK7B4YDThCRrDEjQ-w8kH4OsYZA3QudtisQrMb9Dz5j_uBTr86k5XoqB4mtETm0XRwqTapXSptLGTK_-q74gTjvr</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2235435523</pqid></control><display><type>article</type><title>Towards Task and Architecture-Independent Generalization Gap Predictors</title><source>Free E- Journals</source><creator>Yak, Scott ; Gonzalvo, Javier ; Mazzawi, Hanna</creator><creatorcontrib>Yak, Scott ; Gonzalvo, Javier ; Mazzawi, Hanna</creatorcontrib><description>Can we use deep learning to predict when deep learning works? Our results suggest the affirmative. We created a dataset by training 13,500 neural networks with different architectures, on different variations of spiral datasets, and using different optimization parameters. We used this dataset to train task-independent and architecture-independent generalization gap predictors for those neural networks. We extend Jiang et al. (2018) to also use DNNs and RNNs and show that they outperform the linear model, obtaining \(R^2=0.965\). We also show results for architecture-independent, task-independent, and out-of-distribution generalization gap prediction tasks. Both DNNs and RNNs consistently and significantly outperform linear models, with RNNs obtaining \(R^2=0.584\).</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Architecture ; Machine learning ; Neural networks ; Optimization</subject><ispartof>arXiv.org, 2019-06</ispartof><rights>2019. This work is published under http://creativecommons.org/licenses/by-nc-sa/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>777,781</link.rule.ids></links><search><creatorcontrib>Yak, Scott</creatorcontrib><creatorcontrib>Gonzalvo, Javier</creatorcontrib><creatorcontrib>Mazzawi, Hanna</creatorcontrib><title>Towards Task and Architecture-Independent Generalization Gap Predictors</title><title>arXiv.org</title><description>Can we use deep learning to predict when deep learning works? Our results suggest the affirmative. We created a dataset by training 13,500 neural networks with different architectures, on different variations of spiral datasets, and using different optimization parameters. We used this dataset to train task-independent and architecture-independent generalization gap predictors for those neural networks. We extend Jiang et al. (2018) to also use DNNs and RNNs and show that they outperform the linear model, obtaining \(R^2=0.965\). We also show results for architecture-independent, task-independent, and out-of-distribution generalization gap prediction tasks. Both DNNs and RNNs consistently and significantly outperform linear models, with RNNs obtaining \(R^2=0.584\).</description><subject>Architecture</subject><subject>Machine learning</subject><subject>Neural networks</subject><subject>Optimization</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNjrEKwjAURYMgWLT_EHAu1KRRVxGtbg7dyyN5YmpJ6nspgl9vBz_A5Z7hnOHORKa03hT7SqmFyJm7sizVdqeM0Zmom_gGciwb4KeE4OSB7MMntGkkLK7B4YDThCRrDEjQ-w8kH4OsYZA3QudtisQrMb9Dz5j_uBTr86k5XoqB4mtETm0XRwqTapXSptLGTK_-q74gTjvr</recordid><startdate>20190604</startdate><enddate>20190604</enddate><creator>Yak, Scott</creator><creator>Gonzalvo, Javier</creator><creator>Mazzawi, Hanna</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>20190604</creationdate><title>Towards Task and Architecture-Independent Generalization Gap Predictors</title><author>Yak, Scott ; Gonzalvo, Javier ; Mazzawi, Hanna</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_22354355233</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Architecture</topic><topic>Machine learning</topic><topic>Neural networks</topic><topic>Optimization</topic><toplevel>online_resources</toplevel><creatorcontrib>Yak, Scott</creatorcontrib><creatorcontrib>Gonzalvo, Javier</creatorcontrib><creatorcontrib>Mazzawi, Hanna</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>Yak, Scott</au><au>Gonzalvo, Javier</au><au>Mazzawi, Hanna</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Towards Task and Architecture-Independent Generalization Gap Predictors</atitle><jtitle>arXiv.org</jtitle><date>2019-06-04</date><risdate>2019</risdate><eissn>2331-8422</eissn><abstract>Can we use deep learning to predict when deep learning works? Our results suggest the affirmative. We created a dataset by training 13,500 neural networks with different architectures, on different variations of spiral datasets, and using different optimization parameters. We used this dataset to train task-independent and architecture-independent generalization gap predictors for those neural networks. We extend Jiang et al. (2018) to also use DNNs and RNNs and show that they outperform the linear model, obtaining \(R^2=0.965\). We also show results for architecture-independent, task-independent, and out-of-distribution generalization gap prediction tasks. Both DNNs and RNNs consistently and significantly outperform linear models, with RNNs obtaining \(R^2=0.584\).</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, 2019-06
issn 2331-8422
language eng
recordid cdi_proquest_journals_2235435523
source Free E- Journals
subjects Architecture
Machine learning
Neural networks
Optimization
title Towards Task and Architecture-Independent Generalization Gap Predictors
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-20T12%3A52%3A59IST&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=Towards%20Task%20and%20Architecture-Independent%20Generalization%20Gap%20Predictors&rft.jtitle=arXiv.org&rft.au=Yak,%20Scott&rft.date=2019-06-04&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2235435523%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2235435523&rft_id=info:pmid/&rfr_iscdi=true