{"authors":[{"id":null,"fullName":"De, Abir","name":null,"surname":null,"rank":1,"pid":null},{"id":null,"fullName":"Okati, Nastaran","name":"Nastaran","surname":"Okati","rank":2,"pid":null},{"id":null,"fullName":"Zarezade, Ali","name":"Ali","surname":"Zarezade","rank":3,"pid":null},{"id":null,"fullName":"Gomez Rodriguez, Manuel","name":"Manuel","surname":"Gomez Rodriguez","rank":4,"pid":null}],"openAccessColor":null,"publiclyFunded":false,"eoscIfGuidelines":null,"type":"publication","language":{"code":"und","label":"Undetermined"},"countries":null,"subjects":[{"subject":{"scheme":"keyword","value":"FOS: Computer and information sciences"},"provenance":null},{"subject":{"scheme":"keyword","value":"Computer Science - Machine Learning"},"provenance":null},{"subject":{"scheme":"keyword","value":"Statistics - Machine Learning"},"provenance":null},{"subject":{"scheme":"FOS","value":"0202 electrical engineering, electronic engineering, information engineering"},"provenance":null},{"subject":{"scheme":"keyword","value":"Machine Learning (stat.ML)"},"provenance":null},{"subject":{"scheme":"FOS","value":"02 engineering and technology"},"provenance":null},{"subject":{"scheme":"FOS","value":"01 natural sciences"},"provenance":null},{"subject":{"scheme":"FOS","value":"0105 earth and related environmental sciences"},"provenance":null},{"subject":{"scheme":"keyword","value":"Machine Learning (cs.LG)"},"provenance":null}],"mainTitle":"Classification Under Human Assistance","subTitle":null,"descriptions":["<jats:p>Most supervised learning models are trained for full automation. However, their predictions are sometimes worse than those by human experts on some specific instances. Motivated by this empirical observation, our goal is to design classifiers that are optimized to operate under different automation levels. More specifically, we focus on convex margin-based classifiers and first show that the problem is NP-hard. Then, we further show that, for support vector machines, the corresponding objective function can be expressed as the difference of two functions f = g - c, where g is monotone, non-negative and gamma-weakly submodular, and c is non-negative and modular. This representation allows a recently introduced deterministic greedy algorithm, as well as a more efficient randomized variant of the algorithm, to enjoy approximation guarantees at solving the problem. Experiments on synthetic and real-world data from several applications in medical diagnosis illustrate our theoretical findings and demonstrate that, under human assistance, supervised learning models trained to operate under different automation levels can outperform those trained for full automation as well as humans operating alone.</jats:p>"],"publicationDate":"2021-05-18","publisher":"Association for the Advancement of Artificial Intelligence (AAAI)","embargoEndDate":"2020-06-01","sources":["Crossref","AAAI Technical Track on Humans and AI","Proceedings of the AAAI Conference on Artificial Intelligence"],"formats":null,"contributors":null,"coverages":null,"bestAccessRight":{"code":"c_abf2","label":"OPEN","scheme":"http://vocabularies.coar-repositories.org/documentation/access_rights/"},"container":{"name":"Proceedings of the AAAI Conference on Artificial Intelligence","issnPrinted":"2159-5399","issnOnline":"2374-3468","issnLinking":null,"ep":"5913","iss":null,"sp":"5905","vol":"35","edition":null,"conferencePlace":null,"conferenceDate":null},"documentationUrls":null,"codeRepositoryUrl":null,"programmingLanguage":null,"contactPeople":null,"contactGroups":null,"tools":null,"size":null,"version":null,"geoLocations":null,"id":"doi_dedup___::a1fb558dc3883cfde55bc901cfd0c0e6","originalIds":["10.1609/aaai.v35i7.16738","50|doiboost____|a1fb558dc3883cfde55bc901cfd0c0e6","50|datacite____::8780a0ba92dcc3d5e15c7ff0fdb11c03","10.48550/arxiv.2006.11845","oai:arXiv.org:2006.11845","50|od________18::21bbd3a6cfca1a1b501c4d872ac0264a","50|dblp________::28d9abf2476b1380694510bd7b445c0a","50|dblp________::2e8e3446e852f68d4fb86d7c333731d1","oai:pure.mpg.de:item_3338900","50|od______1874::5bdfffcf840359ea1c9368414640c76d","50|r3c4b2081b22::a1fb558dc3883cfde55bc901cfd0c0e6","50|sygma_______::a1fb558dc3883cfde55bc901cfd0c0e6"],"pids":[{"scheme":"doi","value":"10.1609/aaai.v35i7.16738"},{"scheme":"doi","value":"10.48550/arxiv.2006.11845"},{"scheme":"arXiv","value":"2006.11845"},{"scheme":"handle","value":"21.11116/0000-0009-1A24-2"}],"dateOfCollection":null,"lastUpdateTimeStamp":null,"indicators":{"citationImpact":{"citationCount":13.0,"influence":3.2490097E-9,"popularity":9.608475E-9,"impulse":7.0,"citationClass":"C4","influenceClass":"C4","impulseClass":"C4","popularityClass":"C4"}},"projects":[{"id":"corda__h2020::83d6aa753c4b6e94582f4ef733616892","code":"945719","acronym":"HumanML","title":"Human-Centric Machine Learning","funder":"European Commission","pids":[{"scheme":"doi","value":"10.3030/945719"}]}],"organizations":[{"legalName":"Max Planck Institute for Software Systems","acronym":"MPI-SWS","id":"openorgs____::a6be3b23fdb0e61554b444ec8c1e8d1c","pids":[{"scheme":"Wikidata","value":"Q881238"},{"scheme":"GRID","value":"grid.469860.5"},{"scheme":"ISNI","value":"000000040492020X"},{"scheme":"ROR","value":"https://ror.org/02pe2kf23"},{"scheme":"wikidata","value":"Q881238"}],"countries":[{"code":"DE","label":"Germany"}],"websiteurl":"https://www.mpi-sws.org/index.php"},{"legalName":"Max Planck Society","acronym":"MPG","id":"openorgs____::5a405a89387d1881afa956f475994e10","pids":[{"scheme":"ISNI","value":"0000000121051091"},{"scheme":"ROR","value":"https://ror.org/01hhn8329"},{"scheme":"GRID","value":"grid.4372.2"},{"scheme":"PIC","value":"999990267"},{"scheme":"RRID","value":"RRID:nlx_55250"},{"scheme":"FundRef","value":"501100004189"},{"scheme":"RRID","value":"RRID:SCR_011366"},{"scheme":"Wikidata","value":"Q158085"},{"scheme":"OrgRef","value":"614753"},{"scheme":"mag_id","value":"149899117"},{"scheme":"wikidata","value":"Q158085"},{"scheme":"fundref","value":"501100004189"}],"countries":[{"code":"DE","label":"Germany"}],"websiteurl":"http://www.mpg.de/en"}],"communities":null,"collectedFrom":[{"key":"openaire____::9e3be59865b2c1c335d32dae2fe7b254","value":"Datacite"},{"key":"openaire____::a8db6f6b2ce4fe72e8b2314a9a93e7d9","value":"Sygma"},{"key":"opendoar____::958adb57686c2fdec5796398de5f317a","value":"MPG.PuRe"},{"key":"openaire____::081b82f96300b6a6e3d282bad31cb6e2","value":"Crossref"},{"key":"openaire____::d8b68cc0a53121f6f896883a7d60c1db","value":"DBLP"},{"key":"opendoar____::6f4922f45568161a8cdf4ad2299f6d23","value":"arXiv.org e-Print Archive"},{"key":"re3data_____::c4b2081b224be6b3e79d0e5e5556f631","value":"European Union Open Data Portal"}],"instances":[{"pids":[{"scheme":"doi","value":"10.1609/aaai.v35i7.16738"}],"type":"Article","urls":["https://doi.org/10.1609/aaai.v35i7.16738"],"publicationDate":"2021-05-18","refereed":"peerReviewed","hostedBy":{"key":"issn___print::c3eeaf213793b1a8aa96af199fd54aa4","value":"Proceedings of the AAAI Conference on Artificial Intelligence"},"collectedFrom":{"key":"openaire____::081b82f96300b6a6e3d282bad31cb6e2","value":"Crossref"}},{"pids":[{"scheme":"doi","value":"10.48550/arxiv.2006.11845"}],"license":"arXiv Non-Exclusive Distribution","type":"Article","urls":["https://dx.doi.org/10.48550/arxiv.2006.11845"],"publicationDate":"2020-01-01","refereed":"nonPeerReviewed","hostedBy":{"key":"openaire____::55045bd2a65019fd8e6741a755395c8c","value":"Unknown Repository"},"collectedFrom":{"key":"openaire____::9e3be59865b2c1c335d32dae2fe7b254","value":"Datacite"}},{"pids":[{"scheme":"arXiv","value":"2006.11845"}],"accessRight":{"code":"c_abf2","label":"OPEN","scheme":"http://vocabularies.coar-repositories.org/documentation/access_rights/","openAccessRoute":null},"type":"Preprint","urls":["http://arxiv.org/abs/2006.11845"],"publicationDate":"2020-06-21","refereed":"nonPeerReviewed","hostedBy":{"key":"opendoar____::6f4922f45568161a8cdf4ad2299f6d23","value":"arXiv.org e-Print Archive"},"collectedFrom":{"key":"opendoar____::6f4922f45568161a8cdf4ad2299f6d23","value":"arXiv.org e-Print Archive"}},{"alternateIdentifiers":[{"scheme":"arXiv","value":"2006.11845"}],"type":"Article","urls":["https://arxiv.org/abs/2006.11845","https://dblp.org/rec/journals/corr/abs-2006-11845.html"],"publicationDate":"2020-01-01","refereed":"nonPeerReviewed","hostedBy":{"key":"openaire____::d8b68cc0a53121f6f896883a7d60c1db","value":"DBLP"},"collectedFrom":{"key":"openaire____::d8b68cc0a53121f6f896883a7d60c1db","value":"DBLP"}},{"alternateIdentifiers":[{"scheme":"doi","value":"10.1609/aaai.v35i7.16738"}],"type":"Conference object","urls":["https://dblp.org/rec/conf/aaai/DeOZR21.html","https://doi.org/10.1609/aaai.v35i7.16738"],"refereed":"nonPeerReviewed","hostedBy":{"key":"openaire____::d8b68cc0a53121f6f896883a7d60c1db","value":"DBLP"},"collectedFrom":{"key":"openaire____::d8b68cc0a53121f6f896883a7d60c1db","value":"DBLP"}},{"pids":[{"scheme":"handle","value":"21.11116/0000-0009-1A24-2"}],"alternateIdentifiers":[{"scheme":"urn","value":"https://ojs.aaai.org/index.php/AAAI/article/view/16738"}],"type":"Conference object","urls":["https://hdl.handle.net/21.11116/0000-0009-1A24-2"],"publicationDate":"2021-01-01","refereed":"nonPeerReviewed","hostedBy":{"key":"opendoar____::958adb57686c2fdec5796398de5f317a","value":"MPG.PuRe"},"collectedFrom":{"key":"opendoar____::958adb57686c2fdec5796398de5f317a","value":"MPG.PuRe"}},{"alternateIdentifiers":[{"scheme":"doi","value":"10.1609/aaai.v35i7.16738"}],"type":"Conference object","urls":["http://dx.doi.org/10.1609/aaai.v35i7.16738"],"publicationDate":"2021-01-01","refereed":"nonPeerReviewed","hostedBy":{"key":"openaire____::55045bd2a65019fd8e6741a755395c8c","value":"Unknown Repository"},"collectedFrom":{"key":"re3data_____::c4b2081b224be6b3e79d0e5e5556f631","value":"European Union Open Data Portal"}},{"alternateIdentifiers":[{"scheme":"doi","value":"10.1609/aaai.v35i7.16738"}],"type":"Article","urls":["http://dx.doi.org/10.1609/aaai.v35i7.16738"],"refereed":"nonPeerReviewed","hostedBy":{"key":"issn___print::c3eeaf213793b1a8aa96af199fd54aa4","value":"Proceedings of the AAAI Conference on Artificial Intelligence"},"collectedFrom":{"key":"openaire____::a8db6f6b2ce4fe72e8b2314a9a93e7d9","value":"Sygma"}}],"links":[{"header":{"relationType":"resultProject","relationClass":"isProducedBy","relatedIdentifier":"corda__h2020::83d6aa753c4b6e94582f4ef733616892","relatedRecordType":"project","relationProvenance":"sysimport:crosswalk","trust":"0.9"},"collectedfrom":[{"dsId":"openaire____::a55eb91348674d853191f4f4fd73d078","dsName":"CORDA - COmmon Research DAta Warehouse - Horizon 2020"}],"projectTitle":"Human-Centric Machine Learning","code":"945719","funding":{"funder":{"id":"ec__________::EC","shortname":"EC","name":"European Commission","jurisdiction":{"code":"EU","label":"European Union"},"pid":null},"level0":{"id":"ec__________::EC::H2020","description":"Horizon 2020 Framework Programme","name":"H2020"},"level1":{"id":"ec__________::EC::H2020::ERC","description":"European Research Council","name":"ERC"},"level2":{"id":"ec__________::EC::H2020::ERC::ERC-STG","description":"Starting Grant","name":"ERC-STG"}},"startDate":"2020-11-01","endDate":"2025-10-31"}],"otherTitles":null,"inDiamondJournal":false,"green":true,"isGreen":true,"isInDiamondJournal":false}