Matches in SemOpenAlex for { <https://semopenalex.org/work/W3131465739> ?p ?o ?g. }
- W3131465739 abstract "Self-training is a standard approach to semi-supervised learning where the learner's own predictions on unlabeled data are used as supervision during training. In this paper, we reinterpret this label assignment process as an optimal transportation problem between examples and classes, wherein the cost of assigning an example to a class is mediated by the current predictions of the classifier. This formulation facilitates a practical annealing strategy for label assignment and allows for the inclusion of prior knowledge on class proportions via flexible upper bound constraints. The solutions to these assignment problems can be efficiently approximated using Sinkhorn iteration, thus enabling their use in the inner loop of standard stochastic optimization algorithms. We demonstrate the effectiveness of our algorithm on the CIFAR-10, CIFAR-100, and SVHN datasets in comparison with FixMatch, a state-of-the-art self-training algorithm. Our code is available at this https URL." @default.
- W3131465739 created "2021-03-01" @default.
- W3131465739 creator A5002538469 @default.
- W3131465739 creator A5034622870 @default.
- W3131465739 creator A5079503799 @default.
- W3131465739 date "2021-02-17" @default.
- W3131465739 modified "2023-09-27" @default.
- W3131465739 title "Sinkhorn Label Allocation: Semi-Supervised Classification via Annealed Self-Training" @default.
- W3131465739 cites W1630959083 @default.
- W3131465739 cites W1973117121 @default.
- W3131465739 cites W1975165783 @default.
- W3131465739 cites W2010495720 @default.
- W3131465739 cites W2020018978 @default.
- W3131465739 cites W2036996178 @default.
- W3131465739 cites W2048679005 @default.
- W3131465739 cites W2079057609 @default.
- W3131465739 cites W2088986191 @default.
- W3131465739 cites W2097089247 @default.
- W3131465739 cites W2098136027 @default.
- W3131465739 cites W2101210369 @default.
- W3131465739 cites W2102019796 @default.
- W3131465739 cites W2103284199 @default.
- W3131465739 cites W2107008379 @default.
- W3131465739 cites W2111316763 @default.
- W3131465739 cites W2111557120 @default.
- W3131465739 cites W2117729721 @default.
- W3131465739 cites W2122565017 @default.
- W3131465739 cites W2129068307 @default.
- W3131465739 cites W2132984949 @default.
- W3131465739 cites W2134134392 @default.
- W3131465739 cites W2141416357 @default.
- W3131465739 cites W2145494108 @default.
- W3131465739 cites W2147025064 @default.
- W3131465739 cites W2154368244 @default.
- W3131465739 cites W2158131535 @default.
- W3131465739 cites W2296073425 @default.
- W3131465739 cites W2335728318 @default.
- W3131465739 cites W2531327146 @default.
- W3131465739 cites W2592691248 @default.
- W3131465739 cites W2746314669 @default.
- W3131465739 cites W2910300272 @default.
- W3131465739 cites W2947896848 @default.
- W3131465739 cites W2962369866 @default.
- W3131465739 cites W2962739340 @default.
- W3131465739 cites W2963262394 @default.
- W3131465739 cites W2963435192 @default.
- W3131465739 cites W2964137095 @default.
- W3131465739 cites W2964212410 @default.
- W3131465739 cites W2964324211 @default.
- W3131465739 cites W2978426779 @default.
- W3131465739 cites W2996501936 @default.
- W3131465739 cites W3035160371 @default.
- W3131465739 cites W3095121901 @default.
- W3131465739 cites W3099570996 @default.
- W3131465739 cites W3102631365 @default.
- W3131465739 cites W3118608800 @default.
- W3131465739 hasPublicationYear "2021" @default.
- W3131465739 type Work @default.
- W3131465739 sameAs 3131465739 @default.
- W3131465739 citedByCount "2" @default.
- W3131465739 countsByYear W31314657392021 @default.
- W3131465739 crossrefType "posted-content" @default.
- W3131465739 hasAuthorship W3131465739A5002538469 @default.
- W3131465739 hasAuthorship W3131465739A5034622870 @default.
- W3131465739 hasAuthorship W3131465739A5079503799 @default.
- W3131465739 hasConcept C11413529 @default.
- W3131465739 hasConcept C119857082 @default.
- W3131465739 hasConcept C126255220 @default.
- W3131465739 hasConcept C126980161 @default.
- W3131465739 hasConcept C154945302 @default.
- W3131465739 hasConcept C177264268 @default.
- W3131465739 hasConcept C199360897 @default.
- W3131465739 hasConcept C2776760102 @default.
- W3131465739 hasConcept C2777212361 @default.
- W3131465739 hasConcept C33923547 @default.
- W3131465739 hasConcept C41008148 @default.
- W3131465739 hasConcept C51632099 @default.
- W3131465739 hasConcept C95623464 @default.
- W3131465739 hasConceptScore W3131465739C11413529 @default.
- W3131465739 hasConceptScore W3131465739C119857082 @default.
- W3131465739 hasConceptScore W3131465739C126255220 @default.
- W3131465739 hasConceptScore W3131465739C126980161 @default.
- W3131465739 hasConceptScore W3131465739C154945302 @default.
- W3131465739 hasConceptScore W3131465739C177264268 @default.
- W3131465739 hasConceptScore W3131465739C199360897 @default.
- W3131465739 hasConceptScore W3131465739C2776760102 @default.
- W3131465739 hasConceptScore W3131465739C2777212361 @default.
- W3131465739 hasConceptScore W3131465739C33923547 @default.
- W3131465739 hasConceptScore W3131465739C41008148 @default.
- W3131465739 hasConceptScore W3131465739C51632099 @default.
- W3131465739 hasConceptScore W3131465739C95623464 @default.
- W3131465739 hasLocation W31314657391 @default.
- W3131465739 hasOpenAccess W3131465739 @default.
- W3131465739 hasPrimaryLocation W31314657391 @default.
- W3131465739 hasRelatedWork W1854373118 @default.
- W3131465739 hasRelatedWork W2084507750 @default.
- W3131465739 hasRelatedWork W2140045824 @default.
- W3131465739 hasRelatedWork W2169529361 @default.
- W3131465739 hasRelatedWork W2354047161 @default.
- W3131465739 hasRelatedWork W2382345578 @default.