Matches in SemOpenAlex for { <https://semopenalex.org/work/W4322760400> ?p ?o ?g. }
Showing items 1 to 51 of
51
with 100 items per page.
- W4322760400 abstract "In Multi-Label Learning (MLL), it is extremely challenging to accurately annotate every appearing object due to expensive costs and limited knowledge. When facing such a challenge, a more practical and cheaper alternative should be Single Positive Multi-Label Learning (SPMLL), where only one positive label needs to be provided per sample. Existing SPMLL methods usually assume unknown labels as negatives, which inevitably introduces false negatives as noisy labels. More seriously, Binary Cross Entropy (BCE) loss is often used for training, which is notoriously not robust to noisy labels. To mitigate this issue, we customize an objective function for SPMLL by pushing only one pair of labels apart each time to prevent the domination of negative labels, which is the main culprit of fitting noisy labels in SPMLL. To further combat such noisy labels, we explore the high-rankness of label matrix, which can also push apart different labels. By directly extending from SPMLL to MLL with full labels, a unified loss applicable to both settings is derived. Experiments on real datasets demonstrate that the proposed loss not only performs more robustly to noisy labels for SPMLL but also works well for full labels. Besides, we empirically discover that high-rankness can mitigate the dramatic performance drop in SPMLL. Most surprisingly, even without any regularization or fine-tuned label correction, only adopting our loss defeats state-of-the-art SPMLL methods on CUB, a dataset that severely lacks labels." @default.
- W4322760400 created "2023-03-03" @default.
- W4322760400 creator A5009232538 @default.
- W4322760400 creator A5029012645 @default.
- W4322760400 creator A5083166521 @default.
- W4322760400 date "2023-02-28" @default.
- W4322760400 modified "2023-10-18" @default.
- W4322760400 title "Pushing One Pair of Labels Apart Each Time in Multi-Label Learning: From Single Positive to Full Labels" @default.
- W4322760400 doi "https://doi.org/10.48550/arxiv.2302.14695" @default.
- W4322760400 hasPublicationYear "2023" @default.
- W4322760400 type Work @default.
- W4322760400 citedByCount "0" @default.
- W4322760400 crossrefType "posted-content" @default.
- W4322760400 hasAuthorship W4322760400A5009232538 @default.
- W4322760400 hasAuthorship W4322760400A5029012645 @default.
- W4322760400 hasAuthorship W4322760400A5083166521 @default.
- W4322760400 hasBestOaLocation W43227604001 @default.
- W4322760400 hasConcept C119857082 @default.
- W4322760400 hasConcept C12267149 @default.
- W4322760400 hasConcept C153180895 @default.
- W4322760400 hasConcept C154945302 @default.
- W4322760400 hasConcept C33923547 @default.
- W4322760400 hasConcept C41008148 @default.
- W4322760400 hasConcept C48372109 @default.
- W4322760400 hasConcept C66905080 @default.
- W4322760400 hasConcept C94375191 @default.
- W4322760400 hasConceptScore W4322760400C119857082 @default.
- W4322760400 hasConceptScore W4322760400C12267149 @default.
- W4322760400 hasConceptScore W4322760400C153180895 @default.
- W4322760400 hasConceptScore W4322760400C154945302 @default.
- W4322760400 hasConceptScore W4322760400C33923547 @default.
- W4322760400 hasConceptScore W4322760400C41008148 @default.
- W4322760400 hasConceptScore W4322760400C48372109 @default.
- W4322760400 hasConceptScore W4322760400C66905080 @default.
- W4322760400 hasConceptScore W4322760400C94375191 @default.
- W4322760400 hasLocation W43227604001 @default.
- W4322760400 hasOpenAccess W4322760400 @default.
- W4322760400 hasPrimaryLocation W43227604001 @default.
- W4322760400 hasRelatedWork W2112343299 @default.
- W4322760400 hasRelatedWork W2972984751 @default.
- W4322760400 hasRelatedWork W3133593829 @default.
- W4322760400 hasRelatedWork W4205220472 @default.
- W4322760400 hasRelatedWork W4211088005 @default.
- W4322760400 hasRelatedWork W4285281467 @default.
- W4322760400 hasRelatedWork W4285815787 @default.
- W4322760400 hasRelatedWork W4321442921 @default.
- W4322760400 hasRelatedWork W4322008322 @default.
- W4322760400 hasRelatedWork W2345184372 @default.
- W4322760400 isParatext "false" @default.
- W4322760400 isRetracted "false" @default.
- W4322760400 workType "article" @default.