How to Host a Data Competition: Statistical Advice for Design and Analysis of a Data Competition
Data competitions rely on real-time leaderboards to rank competitor entries and stimulate algorithm improvement. While such competitions have become quite popular and prevalent, particularly in supervised learning formats, their implementations by the host are highly variable. Without careful planni...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2019-01 |
---|---|
Hauptverfasser: | , , , , , |
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 | Anderson-Cook, Christine M Myers, Kary L Lu, Lu Fugate, Michael L Quinlan, Kevin R Pawley, Norma |
description | Data competitions rely on real-time leaderboards to rank competitor entries and stimulate algorithm improvement. While such competitions have become quite popular and prevalent, particularly in supervised learning formats, their implementations by the host are highly variable. Without careful planning, a supervised learning competition is vulnerable to overfitting, where the winning solutions are so closely tuned to the particular set of provided data that they cannot generalize to the underlying problem of interest to the host. This paper outlines some important considerations for strategically designing relevant and informative data sets to maximize the learning outcome from hosting a competition based on our experience. It also describes a post-competition analysis that enables robust and efficient assessment of the strengths and weaknesses of solutions from different competitors, as well as greater understanding of the regions of the input space that are well-solved. The post-competition analysis, which complements the leaderboard, uses exploratory data analysis and generalized linear models (GLMs). The GLMs not only expand the range of results we can explore, they also provide more detailed analysis of individual sub-questions including similarities and differences between algorithms across different types of scenarios, universally easy or hard regions of the input space, and different learning objectives. When coupled with a strategically planned data generation approach, the methods provide richer and more informative summaries to enhance the interpretation of results beyond just the rankings on the leaderboard. The methods are illustrated with a recently completed competition to evaluate algorithms capable of detecting, identifying, and locating radioactive materials in an urban environment. |
doi_str_mv | 10.48550/arxiv.1901.05356 |
format | Article |
fullrecord | <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_1901_05356</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2167973461</sourcerecordid><originalsourceid>FETCH-LOGICAL-a521-3704655f88f3e92eb9a04257b3731fde4ebada54186abfd4fb4fa7e2bbf9bc1f3</originalsourceid><addsrcrecordid>eNptkL1OwzAYRS0kJKrSB2DCEnOCf-OELUqBIFVioHv4nNjIVRqH2C307SktI9Ndjq7uPQjdUJKKXEpyD9O326e0IDQlksvsAs0Y5zTJBWNXaBHChhDCMsWk5DP0XvsvHD2ufYgY8BIi4MpvRxNddH54wG8RogvRtdDjstu71mDrJ7w0wX0MGIYOlwP0h-AC9vafhmt0aaEPZvGXc7R-elxXdbJ6fX6pylUCktGEKyIyKW2eW24KZnQBRDCpNFec2s4Io6EDKWiegbadsFpYUIZpbQvdUsvn6PZce7rfjJPbwnRofjU0Jw1H4u5MjJP_3JkQm43fTcftoWE0U4XiIqP8B0MiYD0</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2167973461</pqid></control><display><type>article</type><title>How to Host a Data Competition: Statistical Advice for Design and Analysis of a Data Competition</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Anderson-Cook, Christine M ; Myers, Kary L ; Lu, Lu ; Fugate, Michael L ; Quinlan, Kevin R ; Pawley, Norma</creator><creatorcontrib>Anderson-Cook, Christine M ; Myers, Kary L ; Lu, Lu ; Fugate, Michael L ; Quinlan, Kevin R ; Pawley, Norma</creatorcontrib><description>Data competitions rely on real-time leaderboards to rank competitor entries and stimulate algorithm improvement. While such competitions have become quite popular and prevalent, particularly in supervised learning formats, their implementations by the host are highly variable. Without careful planning, a supervised learning competition is vulnerable to overfitting, where the winning solutions are so closely tuned to the particular set of provided data that they cannot generalize to the underlying problem of interest to the host. This paper outlines some important considerations for strategically designing relevant and informative data sets to maximize the learning outcome from hosting a competition based on our experience. It also describes a post-competition analysis that enables robust and efficient assessment of the strengths and weaknesses of solutions from different competitors, as well as greater understanding of the regions of the input space that are well-solved. The post-competition analysis, which complements the leaderboard, uses exploratory data analysis and generalized linear models (GLMs). The GLMs not only expand the range of results we can explore, they also provide more detailed analysis of individual sub-questions including similarities and differences between algorithms across different types of scenarios, universally easy or hard regions of the input space, and different learning objectives. When coupled with a strategically planned data generation approach, the methods provide richer and more informative summaries to enhance the interpretation of results beyond just the rankings on the leaderboard. The methods are illustrated with a recently completed competition to evaluate algorithms capable of detecting, identifying, and locating radioactive materials in an urban environment.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.1901.05356</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Competition ; Data analysis ; Generalized linear models ; Radioactive materials ; Statistical models ; Statistics - Applications ; Statistics - Machine Learning ; Supervised learning ; Urban environments</subject><ispartof>arXiv.org, 2019-01</ispartof><rights>2019. 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><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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,780,881,27904</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.1901.05356$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.1002/sam.11404$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Anderson-Cook, Christine M</creatorcontrib><creatorcontrib>Myers, Kary L</creatorcontrib><creatorcontrib>Lu, Lu</creatorcontrib><creatorcontrib>Fugate, Michael L</creatorcontrib><creatorcontrib>Quinlan, Kevin R</creatorcontrib><creatorcontrib>Pawley, Norma</creatorcontrib><title>How to Host a Data Competition: Statistical Advice for Design and Analysis of a Data Competition</title><title>arXiv.org</title><description>Data competitions rely on real-time leaderboards to rank competitor entries and stimulate algorithm improvement. While such competitions have become quite popular and prevalent, particularly in supervised learning formats, their implementations by the host are highly variable. Without careful planning, a supervised learning competition is vulnerable to overfitting, where the winning solutions are so closely tuned to the particular set of provided data that they cannot generalize to the underlying problem of interest to the host. This paper outlines some important considerations for strategically designing relevant and informative data sets to maximize the learning outcome from hosting a competition based on our experience. It also describes a post-competition analysis that enables robust and efficient assessment of the strengths and weaknesses of solutions from different competitors, as well as greater understanding of the regions of the input space that are well-solved. The post-competition analysis, which complements the leaderboard, uses exploratory data analysis and generalized linear models (GLMs). The GLMs not only expand the range of results we can explore, they also provide more detailed analysis of individual sub-questions including similarities and differences between algorithms across different types of scenarios, universally easy or hard regions of the input space, and different learning objectives. When coupled with a strategically planned data generation approach, the methods provide richer and more informative summaries to enhance the interpretation of results beyond just the rankings on the leaderboard. The methods are illustrated with a recently completed competition to evaluate algorithms capable of detecting, identifying, and locating radioactive materials in an urban environment.</description><subject>Algorithms</subject><subject>Competition</subject><subject>Data analysis</subject><subject>Generalized linear models</subject><subject>Radioactive materials</subject><subject>Statistical models</subject><subject>Statistics - Applications</subject><subject>Statistics - Machine Learning</subject><subject>Supervised learning</subject><subject>Urban environments</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><sourceid>GOX</sourceid><recordid>eNptkL1OwzAYRS0kJKrSB2DCEnOCf-OELUqBIFVioHv4nNjIVRqH2C307SktI9Ndjq7uPQjdUJKKXEpyD9O326e0IDQlksvsAs0Y5zTJBWNXaBHChhDCMsWk5DP0XvsvHD2ufYgY8BIi4MpvRxNddH54wG8RogvRtdDjstu71mDrJ7w0wX0MGIYOlwP0h-AC9vafhmt0aaEPZvGXc7R-elxXdbJ6fX6pylUCktGEKyIyKW2eW24KZnQBRDCpNFec2s4Io6EDKWiegbadsFpYUIZpbQvdUsvn6PZce7rfjJPbwnRofjU0Jw1H4u5MjJP_3JkQm43fTcftoWE0U4XiIqP8B0MiYD0</recordid><startdate>20190116</startdate><enddate>20190116</enddate><creator>Anderson-Cook, Christine M</creator><creator>Myers, Kary L</creator><creator>Lu, Lu</creator><creator>Fugate, Michael L</creator><creator>Quinlan, Kevin R</creator><creator>Pawley, Norma</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><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20190116</creationdate><title>How to Host a Data Competition: Statistical Advice for Design and Analysis of a Data Competition</title><author>Anderson-Cook, Christine M ; Myers, Kary L ; Lu, Lu ; Fugate, Michael L ; Quinlan, Kevin R ; Pawley, Norma</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a521-3704655f88f3e92eb9a04257b3731fde4ebada54186abfd4fb4fa7e2bbf9bc1f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Algorithms</topic><topic>Competition</topic><topic>Data analysis</topic><topic>Generalized linear models</topic><topic>Radioactive materials</topic><topic>Statistical models</topic><topic>Statistics - Applications</topic><topic>Statistics - Machine Learning</topic><topic>Supervised learning</topic><topic>Urban environments</topic><toplevel>online_resources</toplevel><creatorcontrib>Anderson-Cook, Christine M</creatorcontrib><creatorcontrib>Myers, Kary L</creatorcontrib><creatorcontrib>Lu, Lu</creatorcontrib><creatorcontrib>Fugate, Michael L</creatorcontrib><creatorcontrib>Quinlan, Kevin R</creatorcontrib><creatorcontrib>Pawley, Norma</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & 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><collection>arXiv Statistics</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Anderson-Cook, Christine M</au><au>Myers, Kary L</au><au>Lu, Lu</au><au>Fugate, Michael L</au><au>Quinlan, Kevin R</au><au>Pawley, Norma</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>How to Host a Data Competition: Statistical Advice for Design and Analysis of a Data Competition</atitle><jtitle>arXiv.org</jtitle><date>2019-01-16</date><risdate>2019</risdate><eissn>2331-8422</eissn><abstract>Data competitions rely on real-time leaderboards to rank competitor entries and stimulate algorithm improvement. While such competitions have become quite popular and prevalent, particularly in supervised learning formats, their implementations by the host are highly variable. Without careful planning, a supervised learning competition is vulnerable to overfitting, where the winning solutions are so closely tuned to the particular set of provided data that they cannot generalize to the underlying problem of interest to the host. This paper outlines some important considerations for strategically designing relevant and informative data sets to maximize the learning outcome from hosting a competition based on our experience. It also describes a post-competition analysis that enables robust and efficient assessment of the strengths and weaknesses of solutions from different competitors, as well as greater understanding of the regions of the input space that are well-solved. The post-competition analysis, which complements the leaderboard, uses exploratory data analysis and generalized linear models (GLMs). The GLMs not only expand the range of results we can explore, they also provide more detailed analysis of individual sub-questions including similarities and differences between algorithms across different types of scenarios, universally easy or hard regions of the input space, and different learning objectives. When coupled with a strategically planned data generation approach, the methods provide richer and more informative summaries to enhance the interpretation of results beyond just the rankings on the leaderboard. The methods are illustrated with a recently completed competition to evaluate algorithms capable of detecting, identifying, and locating radioactive materials in an urban environment.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.1901.05356</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2019-01 |
issn | 2331-8422 |
language | eng |
recordid | cdi_arxiv_primary_1901_05356 |
source | arXiv.org; Free E- Journals |
subjects | Algorithms Competition Data analysis Generalized linear models Radioactive materials Statistical models Statistics - Applications Statistics - Machine Learning Supervised learning Urban environments |
title | How to Host a Data Competition: Statistical Advice for Design and Analysis of a Data Competition |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T19%3A35%3A32IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=How%20to%20Host%20a%20Data%20Competition:%20Statistical%20Advice%20for%20Design%20and%20Analysis%20of%20a%20Data%20Competition&rft.jtitle=arXiv.org&rft.au=Anderson-Cook,%20Christine%20M&rft.date=2019-01-16&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.1901.05356&rft_dat=%3Cproquest_arxiv%3E2167973461%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2167973461&rft_id=info:pmid/&rfr_iscdi=true |