Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285606073> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W4285606073 abstract "Research on human-AI teams usually provides experts with a single label, which ignores the uncertainty in a model's recommendation. Conformal prediction (CP) is a well established line of research that focuses on building a theoretically grounded, calibrated prediction set, which may contain multiple labels. We explore how such prediction sets impact expert decision-making in human-AI teams. Our evaluation on human subjects finds that set valued predictions positively impact experts. However, we notice that the predictive sets provided by CP can be very large, which leads to unhelpful AI assistants. To mitigate this, we introduce D-CP, a method to perform CP on some examples and defer to experts. We prove that D-CP can reduce the prediction set size of non-deferred examples. We show how D-CP performs in quantitative and in human subject experiments (n=120). Our results suggest that CP prediction sets improve human-AI team performance over showing the top-1 prediction alone, and that experts find D-CP prediction sets are more useful than CP prediction sets." @default.
- W4285606073 created "2022-07-16" @default.
- W4285606073 creator A5016469734 @default.
- W4285606073 creator A5042278493 @default.
- W4285606073 creator A5071480038 @default.
- W4285606073 date "2022-07-01" @default.
- W4285606073 modified "2023-10-16" @default.
- W4285606073 title "On the Utility of Prediction Sets in Human-AI Teams" @default.
- W4285606073 doi "https://doi.org/10.24963/ijcai.2022/338" @default.
- W4285606073 hasPublicationYear "2022" @default.
- W4285606073 type Work @default.
- W4285606073 citedByCount "0" @default.
- W4285606073 crossrefType "proceedings-article" @default.
- W4285606073 hasAuthorship W4285606073A5016469734 @default.
- W4285606073 hasAuthorship W4285606073A5042278493 @default.
- W4285606073 hasAuthorship W4285606073A5071480038 @default.
- W4285606073 hasBestOaLocation W42856060731 @default.
- W4285606073 hasConcept C119857082 @default.
- W4285606073 hasConcept C124101348 @default.
- W4285606073 hasConcept C154945302 @default.
- W4285606073 hasConcept C177264268 @default.
- W4285606073 hasConcept C17744445 @default.
- W4285606073 hasConcept C199360897 @default.
- W4285606073 hasConcept C199539241 @default.
- W4285606073 hasConcept C2779913896 @default.
- W4285606073 hasConcept C41008148 @default.
- W4285606073 hasConcept C45804977 @default.
- W4285606073 hasConceptScore W4285606073C119857082 @default.
- W4285606073 hasConceptScore W4285606073C124101348 @default.
- W4285606073 hasConceptScore W4285606073C154945302 @default.
- W4285606073 hasConceptScore W4285606073C177264268 @default.
- W4285606073 hasConceptScore W4285606073C17744445 @default.
- W4285606073 hasConceptScore W4285606073C199360897 @default.
- W4285606073 hasConceptScore W4285606073C199539241 @default.
- W4285606073 hasConceptScore W4285606073C2779913896 @default.
- W4285606073 hasConceptScore W4285606073C41008148 @default.
- W4285606073 hasConceptScore W4285606073C45804977 @default.
- W4285606073 hasLocation W42856060731 @default.
- W4285606073 hasLocation W42856060732 @default.
- W4285606073 hasLocation W42856060733 @default.
- W4285606073 hasOpenAccess W4285606073 @default.
- W4285606073 hasPrimaryLocation W42856060731 @default.
- W4285606073 hasRelatedWork W2961085424 @default.
- W4285606073 hasRelatedWork W3160244858 @default.
- W4285606073 hasRelatedWork W4206988948 @default.
- W4285606073 hasRelatedWork W4221031036 @default.
- W4285606073 hasRelatedWork W4223943233 @default.
- W4285606073 hasRelatedWork W4255117927 @default.
- W4285606073 hasRelatedWork W4306674287 @default.
- W4285606073 hasRelatedWork W4312949351 @default.
- W4285606073 hasRelatedWork W4366376591 @default.
- W4285606073 hasRelatedWork W4224009465 @default.
- W4285606073 isParatext "false" @default.
- W4285606073 isRetracted "false" @default.
- W4285606073 workType "article" @default.