FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2772357980> ?p ?o ?g. }

• W2772357980 abstract "Supervised learning depends on annotated examples, which are taken to be the emph{ground truth}. But these labels often come from noisy crowdsourcing platforms, like Amazon Mechanical Turk. Practitioners typically collect multiple labels per example and aggregate the results to mitigate noise (the classic crowdsourcing problem). Given a fixed annotation budget and unlimited unlabeled data, redundant annotation comes at the expense of fewer labeled examples. This raises two fundamental questions: (1) How can we best learn from noisy workers? (2) How should we allocate our labeling budget to maximize the performance of a classifier? We propose a new algorithm for jointly modeling labels and worker quality from noisy crowd-sourced data. The alternating minimization proceeds in rounds, estimating worker quality from disagreement with the current model and then updating the model by optimizing a loss function that accounts for the current estimate of worker quality. Unlike previous approaches, even with only one annotation per example, our algorithm can estimate worker quality. We establish a generalization error bound for models learned with our algorithm and establish theoretically that it's better to label many examples once (vs less multiply) when worker quality is above a threshold. Experiments conducted on both ImageNet (with simulated noisy workers) and MS-COCO (using the real crowdsourced labels) confirm our algorithm's benefits." @default.
• W2772357980 created "2017-12-22" @default.
• W2772357980 creator A5014498545 @default.
• W2772357980 creator A5029448258 @default.
• W2772357980 creator A5056299576 @default.
• W2772357980 date "2017-12-13" @default.
• W2772357980 modified "2023-09-27" @default.
• W2772357980 title "Learning From Noisy Singly-labeled Data" @default.
• W2772357980 cites W1617082848 @default.
• W2772357980 cites W1814633089 @default.
• W2772357980 cites W1866072925 @default.
• W2772357980 cites W2103490241 @default.
• W2772357980 cites W2108598243 @default.
• W2772357980 cites W2129345386 @default.
• W2772357980 cites W2142518823 @default.
• W2772357980 cites W2144372981 @default.
• W2772357980 cites W2147687736 @default.
• W2772357980 cites W2149273804 @default.
• W2772357980 cites W2152009989 @default.
• W2772357980 cites W2152411181 @default.
• W2772357980 cites W2538903535 @default.
• W2772357980 cites W2604132367 @default.
• W2772357980 cites W2745544219 @default.
• W2772357980 cites W2949312134 @default.
• W2772357980 cites W2964273174 @default.
• W2772357980 cites W3118608800 @default.
• W2772357980 cites W3121928352 @default.
• W2772357980 cites W9014458 @default.
• W2772357980 hasPublicationYear "2017" @default.
• W2772357980 type Work @default.
• W2772357980 sameAs 2772357980 @default.
• W2772357980 citedByCount "32" @default.
• W2772357980 countsByYear W27723579802018 @default.
• W2772357980 countsByYear W27723579802019 @default.
• W2772357980 countsByYear W27723579802020 @default.
• W2772357980 countsByYear W27723579802021 @default.
• W2772357980 crossrefType "posted-content" @default.
• W2772357980 hasAuthorship W2772357980A5014498545 @default.
• W2772357980 hasAuthorship W2772357980A5029448258 @default.
• W2772357980 hasAuthorship W2772357980A5056299576 @default.
• W2772357980 hasConcept C111472728 @default.
• W2772357980 hasConcept C115961682 @default.
• W2772357980 hasConcept C119857082 @default.
• W2772357980 hasConcept C133425853 @default.
• W2772357980 hasConcept C134306372 @default.
• W2772357980 hasConcept C136764020 @default.
• W2772357980 hasConcept C138885662 @default.
• W2772357980 hasConcept C146849305 @default.
• W2772357980 hasConcept C154945302 @default.
• W2772357980 hasConcept C159985019 @default.
• W2772357980 hasConcept C162324750 @default.
• W2772357980 hasConcept C177148314 @default.
• W2772357980 hasConcept C192562407 @default.
• W2772357980 hasConcept C2776145971 @default.
• W2772357980 hasConcept C2776321320 @default.
• W2772357980 hasConcept C2779530757 @default.
• W2772357980 hasConcept C33923547 @default.
• W2772357980 hasConcept C41008148 @default.
• W2772357980 hasConcept C4679612 @default.
• W2772357980 hasConcept C62230096 @default.
• W2772357980 hasConcept C8505890 @default.
• W2772357980 hasConcept C95623464 @default.
• W2772357980 hasConcept C99498987 @default.
• W2772357980 hasConceptScore W2772357980C111472728 @default.
• W2772357980 hasConceptScore W2772357980C115961682 @default.
• W2772357980 hasConceptScore W2772357980C119857082 @default.
• W2772357980 hasConceptScore W2772357980C133425853 @default.
• W2772357980 hasConceptScore W2772357980C134306372 @default.
• W2772357980 hasConceptScore W2772357980C136764020 @default.
• W2772357980 hasConceptScore W2772357980C138885662 @default.
• W2772357980 hasConceptScore W2772357980C146849305 @default.
• W2772357980 hasConceptScore W2772357980C154945302 @default.
• W2772357980 hasConceptScore W2772357980C159985019 @default.
• W2772357980 hasConceptScore W2772357980C162324750 @default.
• W2772357980 hasConceptScore W2772357980C177148314 @default.
• W2772357980 hasConceptScore W2772357980C192562407 @default.
• W2772357980 hasConceptScore W2772357980C2776145971 @default.
• W2772357980 hasConceptScore W2772357980C2776321320 @default.
• W2772357980 hasConceptScore W2772357980C2779530757 @default.
• W2772357980 hasConceptScore W2772357980C33923547 @default.
• W2772357980 hasConceptScore W2772357980C41008148 @default.
• W2772357980 hasConceptScore W2772357980C4679612 @default.
• W2772357980 hasConceptScore W2772357980C62230096 @default.
• W2772357980 hasConceptScore W2772357980C8505890 @default.
• W2772357980 hasConceptScore W2772357980C95623464 @default.
• W2772357980 hasConceptScore W2772357980C99498987 @default.
• W2772357980 hasLocation W27723579801 @default.
• W2772357980 hasOpenAccess W2772357980 @default.
• W2772357980 hasPrimaryLocation W27723579801 @default.
• W2772357980 hasRelatedWork W1514928307 @default.
• W2772357980 hasRelatedWork W1866072925 @default.
• W2772357980 hasRelatedWork W1921293667 @default.
• W2772357980 hasRelatedWork W1975128126 @default.
• W2772357980 hasRelatedWork W2108598243 @default.
• W2772357980 hasRelatedWork W2113290770 @default.
• W2772357980 hasRelatedWork W2121056381 @default.
• W2772357980 hasRelatedWork W2149273804 @default.
• W2772357980 hasRelatedWork W2194775991 @default.
• W2772357980 hasRelatedWork W2618574054 @default.
• W2772357980 hasRelatedWork W2752971446 @default.

• next