Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136604105> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W3136604105 abstract "The recently proposed Temporal Ensembling has achieved state-of-the-art results in several semi-supervised learning benchmarks. It maintains an exponential moving average of label predictions on each training example, and penalizes predictions that are inconsistent with this target. However, because the targets change only once per epoch, Temporal Ensembling becomes unwieldy when learning large datasets. To overcome this problem, we propose Mean Teacher, a method that averages model weights instead of label predictions. As an additional benefit, Mean Teacher improves test accuracy and enables training with fewer labels than Temporal Ensembling. Without changing the network architecture, Mean Teacher achieves an error rate of 4.35% on SVHN with 250 labels, outperforming Temporal Ensembling trained with 1000 labels. We also show that a good network architecture is crucial to performance. Combining Mean Teacher and Residual Networks, we improve the state of the art on CIFAR-10 with 4000 labels from 10.55% to 6.28%, and on ImageNet 2012 with 10% of the labels from 35.24% to 9.11%." @default.
- W3136604105 created "2021-03-29" @default.
- W3136604105 creator A5042259625 @default.
- W3136604105 creator A5050971487 @default.
- W3136604105 date "2017-01-01" @default.
- W3136604105 modified "2023-09-24" @default.
- W3136604105 title "Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results." @default.
- W3136604105 hasPublicationYear "2017" @default.
- W3136604105 type Work @default.
- W3136604105 sameAs 3136604105 @default.
- W3136604105 citedByCount "12" @default.
- W3136604105 countsByYear W31366041052020 @default.
- W3136604105 countsByYear W31366041052021 @default.
- W3136604105 crossrefType "proceedings-article" @default.
- W3136604105 hasAuthorship W3136604105A5042259625 @default.
- W3136604105 hasAuthorship W3136604105A5050971487 @default.
- W3136604105 hasConcept C11413529 @default.
- W3136604105 hasConcept C119857082 @default.
- W3136604105 hasConcept C154945302 @default.
- W3136604105 hasConcept C155512373 @default.
- W3136604105 hasConcept C2776436953 @default.
- W3136604105 hasConcept C41008148 @default.
- W3136604105 hasConceptScore W3136604105C11413529 @default.
- W3136604105 hasConceptScore W3136604105C119857082 @default.
- W3136604105 hasConceptScore W3136604105C154945302 @default.
- W3136604105 hasConceptScore W3136604105C155512373 @default.
- W3136604105 hasConceptScore W3136604105C2776436953 @default.
- W3136604105 hasConceptScore W3136604105C41008148 @default.
- W3136604105 hasLocation W31366041051 @default.
- W3136604105 hasOpenAccess W3136604105 @default.
- W3136604105 hasPrimaryLocation W31366041051 @default.
- W3136604105 hasRelatedWork W2118858186 @default.
- W3136604105 hasRelatedWork W2145494108 @default.
- W3136604105 hasRelatedWork W2194775991 @default.
- W3136604105 hasRelatedWork W2645998928 @default.
- W3136604105 hasRelatedWork W2766289485 @default.
- W3136604105 hasRelatedWork W2883725317 @default.
- W3136604105 hasRelatedWork W2901436880 @default.
- W3136604105 hasRelatedWork W2953070460 @default.
- W3136604105 hasRelatedWork W2962369866 @default.
- W3136604105 hasRelatedWork W2962742544 @default.
- W3136604105 hasRelatedWork W2963341956 @default.
- W3136604105 hasRelatedWork W2963399829 @default.
- W3136604105 hasRelatedWork W2964159205 @default.
- W3136604105 hasRelatedWork W2978426779 @default.
- W3136604105 hasRelatedWork W2978471258 @default.
- W3136604105 hasRelatedWork W2990500698 @default.
- W3136604105 hasRelatedWork W2996501936 @default.
- W3136604105 hasRelatedWork W3118608800 @default.
- W3136604105 hasRelatedWork W3205361606 @default.
- W3136604105 hasRelatedWork W2530816535 @default.
- W3136604105 isParatext "false" @default.
- W3136604105 isRetracted "false" @default.
- W3136604105 magId "3136604105" @default.
- W3136604105 workType "article" @default.