Matches in SemOpenAlex for { <https://semopenalex.org/work/W4376133361> ?p ?o ?g. }
- W4376133361 abstract "As governments and industry turn to increased use of automated decision systems, it becomes essential to consider how closely such systems can reproduce human judgment. We identify a core potential failure, finding that annotators label objects differently depending on whether they are being asked a factual question or a normative question. This challenges a natural assumption maintained in many standard machine-learning (ML) data acquisition procedures: that there is no difference between predicting the factual classification of an object and an exercise of judgment about whether an object violates a rule premised on those facts. We find that using factual labels to train models intended for normative judgments introduces a notable measurement error. We show that models trained using factual labels yield significantly different judgments than those trained using normative labels and that the impact of this effect on model performance can exceed that of other factors (e.g., dataset size) that routinely attract attention from ML researchers and practitioners." @default.
- W4376133361 created "2023-05-12" @default.
- W4376133361 creator A5000711679 @default.
- W4376133361 creator A5010008842 @default.
- W4376133361 creator A5018327012 @default.
- W4376133361 creator A5052778643 @default.
- W4376133361 creator A5070063054 @default.
- W4376133361 creator A5076757561 @default.
- W4376133361 date "2023-05-10" @default.
- W4376133361 modified "2023-09-30" @default.
- W4376133361 title "Judging facts, judging norms: Training machine learning models to judge humans requires a modified approach to labeling data" @default.
- W4376133361 cites W12634471 @default.
- W4376133361 cites W146395692 @default.
- W4376133361 cites W1497131193 @default.
- W4376133361 cites W1834627138 @default.
- W4376133361 cites W2038901583 @default.
- W4376133361 cites W2059621286 @default.
- W4376133361 cites W2108598243 @default.
- W4376133361 cites W2117065922 @default.
- W4376133361 cites W2149688032 @default.
- W4376133361 cites W2194775991 @default.
- W4376133361 cites W2332094122 @default.
- W4376133361 cites W2471768434 @default.
- W4376133361 cites W2501134159 @default.
- W4376133361 cites W2531079776 @default.
- W4376133361 cites W2531634031 @default.
- W4376133361 cites W2551317447 @default.
- W4376133361 cites W2736351703 @default.
- W4376133361 cites W2963846996 @default.
- W4376133361 cites W2964292098 @default.
- W4376133361 cites W2971560587 @default.
- W4376133361 cites W3016154458 @default.
- W4376133361 cites W3017195291 @default.
- W4376133361 cites W3033733989 @default.
- W4376133361 cites W3044667543 @default.
- W4376133361 cites W3100279624 @default.
- W4376133361 cites W3104041537 @default.
- W4376133361 cites W3123668717 @default.
- W4376133361 cites W3130937151 @default.
- W4376133361 cites W3134508370 @default.
- W4376133361 cites W3161159443 @default.
- W4376133361 cites W3163411042 @default.
- W4376133361 cites W3163443091 @default.
- W4376133361 cites W3184924454 @default.
- W4376133361 cites W3187543315 @default.
- W4376133361 cites W3212368439 @default.
- W4376133361 cites W3212464620 @default.
- W4376133361 cites W4242981563 @default.
- W4376133361 cites W4244135171 @default.
- W4376133361 cites W4283156811 @default.
- W4376133361 cites W4309651064 @default.
- W4376133361 doi "https://doi.org/10.1126/sciadv.abq0701" @default.
- W4376133361 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37163590" @default.
- W4376133361 hasPublicationYear "2023" @default.
- W4376133361 type Work @default.
- W4376133361 citedByCount "0" @default.
- W4376133361 crossrefType "journal-article" @default.
- W4376133361 hasAuthorship W4376133361A5000711679 @default.
- W4376133361 hasAuthorship W4376133361A5010008842 @default.
- W4376133361 hasAuthorship W4376133361A5018327012 @default.
- W4376133361 hasAuthorship W4376133361A5052778643 @default.
- W4376133361 hasAuthorship W4376133361A5070063054 @default.
- W4376133361 hasAuthorship W4376133361A5076757561 @default.
- W4376133361 hasBestOaLocation W43761333611 @default.
- W4376133361 hasConcept C111472728 @default.
- W4376133361 hasConcept C119857082 @default.
- W4376133361 hasConcept C137549413 @default.
- W4376133361 hasConcept C138885662 @default.
- W4376133361 hasConcept C154945302 @default.
- W4376133361 hasConcept C15744967 @default.
- W4376133361 hasConcept C180747234 @default.
- W4376133361 hasConcept C2164484 @default.
- W4376133361 hasConcept C2781238097 @default.
- W4376133361 hasConcept C41008148 @default.
- W4376133361 hasConcept C44725695 @default.
- W4376133361 hasConcept C76155785 @default.
- W4376133361 hasConceptScore W4376133361C111472728 @default.
- W4376133361 hasConceptScore W4376133361C119857082 @default.
- W4376133361 hasConceptScore W4376133361C137549413 @default.
- W4376133361 hasConceptScore W4376133361C138885662 @default.
- W4376133361 hasConceptScore W4376133361C154945302 @default.
- W4376133361 hasConceptScore W4376133361C15744967 @default.
- W4376133361 hasConceptScore W4376133361C180747234 @default.
- W4376133361 hasConceptScore W4376133361C2164484 @default.
- W4376133361 hasConceptScore W4376133361C2781238097 @default.
- W4376133361 hasConceptScore W4376133361C41008148 @default.
- W4376133361 hasConceptScore W4376133361C44725695 @default.
- W4376133361 hasConceptScore W4376133361C76155785 @default.
- W4376133361 hasIssue "19" @default.
- W4376133361 hasLocation W43761333611 @default.
- W4376133361 hasLocation W43761333612 @default.
- W4376133361 hasOpenAccess W4376133361 @default.
- W4376133361 hasPrimaryLocation W43761333611 @default.
- W4376133361 hasRelatedWork W1975684773 @default.
- W4376133361 hasRelatedWork W2066059397 @default.
- W4376133361 hasRelatedWork W2091129439 @default.
- W4376133361 hasRelatedWork W2748952813 @default.
- W4376133361 hasRelatedWork W2774806438 @default.
- W4376133361 hasRelatedWork W2899084033 @default.
- W4376133361 hasRelatedWork W2961085424 @default.