Matches in SemOpenAlex for { <https://semopenalex.org/work/W2904708794> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W2904708794 abstract "We consider the task of training classifiers without labels. We propose a weakly supervised method---adversarial label learning---that trains classifiers to perform well against an adversary that chooses labels for training data. The weak supervision constrains what labels the adversary can choose. The method therefore minimizes an upper bound of the classifier's error rate using projected primal-dual subgradient descent. Minimizing this bound protects against bias and dependencies in the weak supervision. Experiments on three real datasets show that our method can train without labels and outperforms other approaches for weakly supervised learning." @default.
- W2904708794 created "2018-12-22" @default.
- W2904708794 creator A5002044887 @default.
- W2904708794 creator A5088795873 @default.
- W2904708794 date "2018-05-22" @default.
- W2904708794 modified "2023-09-23" @default.
- W2904708794 title "Adversarial Label Learning" @default.
- W2904708794 doi "https://doi.org/10.48550/arxiv.1805.08877" @default.
- W2904708794 hasPublicationYear "2018" @default.
- W2904708794 type Work @default.
- W2904708794 sameAs 2904708794 @default.
- W2904708794 citedByCount "0" @default.
- W2904708794 crossrefType "posted-content" @default.
- W2904708794 hasAuthorship W2904708794A5002044887 @default.
- W2904708794 hasAuthorship W2904708794A5088795873 @default.
- W2904708794 hasBestOaLocation W29047087941 @default.
- W2904708794 hasConcept C119857082 @default.
- W2904708794 hasConcept C134306372 @default.
- W2904708794 hasConcept C154945302 @default.
- W2904708794 hasConcept C158968445 @default.
- W2904708794 hasConcept C162324750 @default.
- W2904708794 hasConcept C187736073 @default.
- W2904708794 hasConcept C2776145971 @default.
- W2904708794 hasConcept C2780451532 @default.
- W2904708794 hasConcept C33923547 @default.
- W2904708794 hasConcept C37736160 @default.
- W2904708794 hasConcept C38652104 @default.
- W2904708794 hasConcept C41008148 @default.
- W2904708794 hasConcept C41065033 @default.
- W2904708794 hasConcept C77553402 @default.
- W2904708794 hasConcept C95623464 @default.
- W2904708794 hasConceptScore W2904708794C119857082 @default.
- W2904708794 hasConceptScore W2904708794C134306372 @default.
- W2904708794 hasConceptScore W2904708794C154945302 @default.
- W2904708794 hasConceptScore W2904708794C158968445 @default.
- W2904708794 hasConceptScore W2904708794C162324750 @default.
- W2904708794 hasConceptScore W2904708794C187736073 @default.
- W2904708794 hasConceptScore W2904708794C2776145971 @default.
- W2904708794 hasConceptScore W2904708794C2780451532 @default.
- W2904708794 hasConceptScore W2904708794C33923547 @default.
- W2904708794 hasConceptScore W2904708794C37736160 @default.
- W2904708794 hasConceptScore W2904708794C38652104 @default.
- W2904708794 hasConceptScore W2904708794C41008148 @default.
- W2904708794 hasConceptScore W2904708794C41065033 @default.
- W2904708794 hasConceptScore W2904708794C77553402 @default.
- W2904708794 hasConceptScore W2904708794C95623464 @default.
- W2904708794 hasLocation W29047087941 @default.
- W2904708794 hasOpenAccess W2904708794 @default.
- W2904708794 hasPrimaryLocation W29047087941 @default.
- W2904708794 hasRelatedWork W2460937040 @default.
- W2904708794 hasRelatedWork W2804927400 @default.
- W2904708794 hasRelatedWork W2904708794 @default.
- W2904708794 hasRelatedWork W2952541330 @default.
- W2904708794 hasRelatedWork W2953083558 @default.
- W2904708794 hasRelatedWork W2961085424 @default.
- W2904708794 hasRelatedWork W2963444698 @default.
- W2904708794 hasRelatedWork W3037770290 @default.
- W2904708794 hasRelatedWork W4287869347 @default.
- W2904708794 hasRelatedWork W4300511536 @default.
- W2904708794 isParatext "false" @default.
- W2904708794 isRetracted "false" @default.
- W2904708794 magId "2904708794" @default.
- W2904708794 workType "article" @default.