Matches in SemOpenAlex for { <https://semopenalex.org/work/W2892005688> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W2892005688 abstract "Noisy labels modeling makes a convolutional neural network (CNN) more robust for the image classification problem. However, current noisy labels modeling methods usually require an expectation-maximization (EM) based procedure to optimize the parameters, which is computationally expensive. In this paper, we utilize a fast annealing training method to speed up the CNN training in every M-step. Since the training is repeated executed along the entire EM optimization path and obtain many local minimal CNN models from every training cycle, we name it as the Cyclic Annealing Training (CAT) approach. In addition to reducing the training time, CAT can further bagging all the local minimal CNN models at the test time to improve the performance of classification. We evaluate the proposed method on several image classification datasets with different noisy labels patterns, and the results show that our CAT approach outperforms state-of-the-art noisy labels modeling methods." @default.
- W2892005688 created "2018-09-27" @default.
- W2892005688 creator A5017567548 @default.
- W2892005688 creator A5023762528 @default.
- W2892005688 creator A5034104790 @default.
- W2892005688 creator A5047134950 @default.
- W2892005688 creator A5059534557 @default.
- W2892005688 date "2018-10-01" @default.
- W2892005688 modified "2023-09-27" @default.
- W2892005688 title "Cyclic Annealing Training Convolutional Neural Networks for Image Classification with Noisy Labels" @default.
- W2892005688 cites W1605688901 @default.
- W2892005688 cites W1768290534 @default.
- W2892005688 cites W1994550352 @default.
- W2892005688 cites W2012243161 @default.
- W2892005688 cites W2112076978 @default.
- W2892005688 cites W2113290770 @default.
- W2892005688 cites W2138507544 @default.
- W2892005688 cites W2167460663 @default.
- W2892005688 cites W2194775991 @default.
- W2892005688 cites W2252268321 @default.
- W2892005688 cites W2403681572 @default.
- W2892005688 cites W2492899067 @default.
- W2892005688 cites W2612983688 @default.
- W2892005688 cites W2912934387 @default.
- W2892005688 cites W2962762541 @default.
- W2892005688 cites W2963263347 @default.
- W2892005688 cites W2963446085 @default.
- W2892005688 cites W2963446712 @default.
- W2892005688 cites W2973562770 @default.
- W2892005688 cites W3104436273 @default.
- W2892005688 cites W3118608800 @default.
- W2892005688 cites W605727707 @default.
- W2892005688 cites W62203920 @default.
- W2892005688 doi "https://doi.org/10.1109/icip.2018.8451331" @default.
- W2892005688 hasPublicationYear "2018" @default.
- W2892005688 type Work @default.
- W2892005688 sameAs 2892005688 @default.
- W2892005688 citedByCount "3" @default.
- W2892005688 countsByYear W28920056882019 @default.
- W2892005688 crossrefType "proceedings-article" @default.
- W2892005688 hasAuthorship W2892005688A5017567548 @default.
- W2892005688 hasAuthorship W2892005688A5023762528 @default.
- W2892005688 hasAuthorship W2892005688A5034104790 @default.
- W2892005688 hasAuthorship W2892005688A5047134950 @default.
- W2892005688 hasAuthorship W2892005688A5059534557 @default.
- W2892005688 hasConcept C115961682 @default.
- W2892005688 hasConcept C119857082 @default.
- W2892005688 hasConcept C126255220 @default.
- W2892005688 hasConcept C126980161 @default.
- W2892005688 hasConcept C153180895 @default.
- W2892005688 hasConcept C154945302 @default.
- W2892005688 hasConcept C2776330181 @default.
- W2892005688 hasConcept C33923547 @default.
- W2892005688 hasConcept C41008148 @default.
- W2892005688 hasConcept C50644808 @default.
- W2892005688 hasConcept C75294576 @default.
- W2892005688 hasConcept C81363708 @default.
- W2892005688 hasConceptScore W2892005688C115961682 @default.
- W2892005688 hasConceptScore W2892005688C119857082 @default.
- W2892005688 hasConceptScore W2892005688C126255220 @default.
- W2892005688 hasConceptScore W2892005688C126980161 @default.
- W2892005688 hasConceptScore W2892005688C153180895 @default.
- W2892005688 hasConceptScore W2892005688C154945302 @default.
- W2892005688 hasConceptScore W2892005688C2776330181 @default.
- W2892005688 hasConceptScore W2892005688C33923547 @default.
- W2892005688 hasConceptScore W2892005688C41008148 @default.
- W2892005688 hasConceptScore W2892005688C50644808 @default.
- W2892005688 hasConceptScore W2892005688C75294576 @default.
- W2892005688 hasConceptScore W2892005688C81363708 @default.
- W2892005688 hasLocation W28920056881 @default.
- W2892005688 hasOpenAccess W2892005688 @default.
- W2892005688 hasPrimaryLocation W28920056881 @default.
- W2892005688 hasRelatedWork W2406522397 @default.
- W2892005688 hasRelatedWork W2588607706 @default.
- W2892005688 hasRelatedWork W2760944304 @default.
- W2892005688 hasRelatedWork W2767651786 @default.
- W2892005688 hasRelatedWork W2800691917 @default.
- W2892005688 hasRelatedWork W2885073348 @default.
- W2892005688 hasRelatedWork W2955667634 @default.
- W2892005688 hasRelatedWork W2972069047 @default.
- W2892005688 hasRelatedWork W2980720988 @default.
- W2892005688 hasRelatedWork W4200390792 @default.
- W2892005688 isParatext "false" @default.
- W2892005688 isRetracted "false" @default.
- W2892005688 magId "2892005688" @default.
- W2892005688 workType "article" @default.