Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels
Crowdsourcing has become a primary means for label collection in many real-world machine learning applications. A classical method for inferring the true labels from the noisy labels provided by crowdsourcing workers is Dawid-Skene estimator. In this paper, we prove convergence rates of a projected...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| 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 | Gao, Chao Zhou, Dengyong |
| description | Crowdsourcing has become a primary means for label collection in many
real-world machine learning applications. A classical method for inferring the
true labels from the noisy labels provided by crowdsourcing workers is
Dawid-Skene estimator. In this paper, we prove convergence rates of a projected
EM algorithm for the Dawid-Skene estimator. The revealed exponent in the rate
of convergence is shown to be optimal via a lower bound argument. Our work
resolves the long standing issue of whether Dawid-Skene estimator has sound
theoretical guarantees besides its good performance observed in practice. In
addition, a comparative study with majority voting illustrates both advantages
and pitfalls of the Dawid-Skene estimator. |
| doi_str_mv | 10.48550/arxiv.1310.5764 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1310_5764</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1310_5764</sourcerecordid><originalsourceid>FETCH-LOGICAL-a654-424fabe7c2c4c9bf4600f5288499cf08840453cdf266f7cb0e20653cf2d2e76d3</originalsourceid><addsrcrecordid>eNotj1FLwzAUhfPig0zffZL8gc4sTdL2UcqcQmUgxdeSJrmz0CXjpp3z35vqnj4453I5HyEPG7YWpZTsSeNlOK83eQpkocQt-Xwf_HDUF7o_TYkjrYM_Ozw4bxz90JOLFALSbVzaafAHusMwe0tbnKcvChiOtMbwbWOY0ThLG927Md6RG9BjdPdXrkj7sm3r16zZ797q5ybTSopMcAHpvDDcCFP1IBRjIHlZiqoywBKZkLmxwJWCwvTMcaZSANxyVyibr8jj_9s_r-6EaST-dItft_jlvww3S68</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels</title><source>arXiv.org</source><creator>Gao, Chao ; Zhou, Dengyong</creator><creatorcontrib>Gao, Chao ; Zhou, Dengyong</creatorcontrib><description>Crowdsourcing has become a primary means for label collection in many
real-world machine learning applications. A classical method for inferring the
true labels from the noisy labels provided by crowdsourcing workers is
Dawid-Skene estimator. In this paper, we prove convergence rates of a projected
EM algorithm for the Dawid-Skene estimator. The revealed exponent in the rate
of convergence is shown to be optimal via a lower bound argument. Our work
resolves the long standing issue of whether Dawid-Skene estimator has sound
theoretical guarantees besides its good performance observed in practice. In
addition, a comparative study with majority voting illustrates both advantages
and pitfalls of the Dawid-Skene estimator.</description><identifier>DOI: 10.48550/arxiv.1310.5764</identifier><language>eng</language><subject>Mathematics - Statistics Theory ; Statistics - Machine Learning ; Statistics - Theory</subject><creationdate>2013-10</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/1310.5764$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1310.5764$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Gao, Chao</creatorcontrib><creatorcontrib>Zhou, Dengyong</creatorcontrib><title>Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels</title><description>Crowdsourcing has become a primary means for label collection in many
real-world machine learning applications. A classical method for inferring the
true labels from the noisy labels provided by crowdsourcing workers is
Dawid-Skene estimator. In this paper, we prove convergence rates of a projected
EM algorithm for the Dawid-Skene estimator. The revealed exponent in the rate
of convergence is shown to be optimal via a lower bound argument. Our work
resolves the long standing issue of whether Dawid-Skene estimator has sound
theoretical guarantees besides its good performance observed in practice. In
addition, a comparative study with majority voting illustrates both advantages
and pitfalls of the Dawid-Skene estimator.</description><subject>Mathematics - Statistics Theory</subject><subject>Statistics - Machine Learning</subject><subject>Statistics - Theory</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2013</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj1FLwzAUhfPig0zffZL8gc4sTdL2UcqcQmUgxdeSJrmz0CXjpp3z35vqnj4453I5HyEPG7YWpZTsSeNlOK83eQpkocQt-Xwf_HDUF7o_TYkjrYM_Ozw4bxz90JOLFALSbVzaafAHusMwe0tbnKcvChiOtMbwbWOY0ThLG927Md6RG9BjdPdXrkj7sm3r16zZ797q5ybTSopMcAHpvDDcCFP1IBRjIHlZiqoywBKZkLmxwJWCwvTMcaZSANxyVyibr8jj_9s_r-6EaST-dItft_jlvww3S68</recordid><startdate>20131021</startdate><enddate>20131021</enddate><creator>Gao, Chao</creator><creator>Zhou, Dengyong</creator><scope>AKZ</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20131021</creationdate><title>Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels</title><author>Gao, Chao ; Zhou, Dengyong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a654-424fabe7c2c4c9bf4600f5288499cf08840453cdf266f7cb0e20653cf2d2e76d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Mathematics - Statistics Theory</topic><topic>Statistics - Machine Learning</topic><topic>Statistics - Theory</topic><toplevel>online_resources</toplevel><creatorcontrib>Gao, Chao</creatorcontrib><creatorcontrib>Zhou, Dengyong</creatorcontrib><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>Gao, Chao</au><au>Zhou, Dengyong</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels</atitle><date>2013-10-21</date><risdate>2013</risdate><abstract>Crowdsourcing has become a primary means for label collection in many
real-world machine learning applications. A classical method for inferring the
true labels from the noisy labels provided by crowdsourcing workers is
Dawid-Skene estimator. In this paper, we prove convergence rates of a projected
EM algorithm for the Dawid-Skene estimator. The revealed exponent in the rate
of convergence is shown to be optimal via a lower bound argument. Our work
resolves the long standing issue of whether Dawid-Skene estimator has sound
theoretical guarantees besides its good performance observed in practice. In
addition, a comparative study with majority voting illustrates both advantages
and pitfalls of the Dawid-Skene estimator.</abstract><doi>10.48550/arxiv.1310.5764</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.1310.5764 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_1310_5764 |
| source | arXiv.org |
| subjects | Mathematics - Statistics Theory Statistics - Machine Learning Statistics - Theory |
| title | Minimax Optimal Convergence Rates for Estimating Ground Truth from Crowdsourced Labels |
| 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%3A24%3A32IST&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=Minimax%20Optimal%20Convergence%20Rates%20for%20Estimating%20Ground%20Truth%20from%20Crowdsourced%20Labels&rft.au=Gao,%20Chao&rft.date=2013-10-21&rft_id=info:doi/10.48550/arxiv.1310.5764&rft_dat=%3Carxiv_GOX%3E1310_5764%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 |