FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2803154237> ?p ?o ?g. }

• W2803154237 abstract "A major impediment to the application of deep learning to real-world problems is the scarcity of labeled data. Small training sets are in fact of no use to deep networks as, due to the large number of trainable parameters, they will very likely be subject to overfitting phenomena. On the other hand, the increment of the training set size through further manual or semi-automatic labellings can be costly, if not possible at times. Thus, the standard techniques to address this issue are transfer learning and data augmentation, which consists of applying some sort of to existing labeled instances to let the training set grow in size. Although this approach works well in applications such as image classification, where it is relatively simple to design suitable transformation operators, it is not obvious how to apply it in more structured scenarios. Motivated by the observation that in virtually all application domains it is easy to obtain unlabeled data, in this paper we take a different perspective and propose a emph{label augmentation} approach. We start from a small, curated labeled dataset and let the labels propagate through a larger set of unlabeled data using graph transduction techniques. This allows us to naturally use (second-order) similarity information which resides in the data, a source of information which is typically neglected by standard augmentation techniques. In particular, we show that by using known game theoretic transductive processes we can create larger and accurate enough labeled datasets which use results in better trained neural networks. Preliminary experiments are reported which demonstrate a consistent improvement over standard image classification datasets." @default.
• W2803154237 created "2018-06-01" @default.
• W2803154237 creator A5054839433 @default.
• W2803154237 creator A5055143299 @default.
• W2803154237 creator A5079235359 @default.
• W2803154237 creator A5088359990 @default.
• W2803154237 date "2018-05-26" @default.
• W2803154237 modified "2023-09-27" @default.
• W2803154237 title "Transductive Label Augmentation for Improved Deep Network Learning" @default.
• W2803154237 cites W1497443639 @default.
• W2803154237 cites W1522301498 @default.
• W2803154237 cites W1549847953 @default.
• W2803154237 cites W1576445103 @default.
• W2803154237 cites W1944148361 @default.
• W2803154237 cites W1960351623 @default.
• W2803154237 cites W2076063813 @default.
• W2803154237 cites W2091987367 @default.
• W2803154237 cites W2092245690 @default.
• W2803154237 cites W2097117768 @default.
• W2803154237 cites W2112796928 @default.
• W2803154237 cites W2147800946 @default.
• W2803154237 cites W2148603752 @default.
• W2803154237 cites W2151103935 @default.
• W2803154237 cites W2152161678 @default.
• W2803154237 cites W2152322845 @default.
• W2803154237 cites W2156726056 @default.
• W2803154237 cites W2194775991 @default.
• W2803154237 cites W2330024298 @default.
• W2803154237 cites W2522991245 @default.
• W2803154237 cites W2949092679 @default.
• W2803154237 cites W2949416428 @default.
• W2803154237 cites W2949667497 @default.
• W2803154237 cites W2963446712 @default.
• W2803154237 cites W3118608800 @default.
• W2803154237 cites W3122924345 @default.
• W2803154237 hasPublicationYear "2018" @default.
• W2803154237 type Work @default.
• W2803154237 sameAs 2803154237 @default.
• W2803154237 citedByCount "8" @default.
• W2803154237 countsByYear W28031542372018 @default.
• W2803154237 countsByYear W28031542372019 @default.
• W2803154237 countsByYear W28031542372020 @default.
• W2803154237 countsByYear W28031542372021 @default.
• W2803154237 crossrefType "posted-content" @default.
• W2803154237 hasAuthorship W2803154237A5054839433 @default.
• W2803154237 hasAuthorship W2803154237A5055143299 @default.
• W2803154237 hasAuthorship W2803154237A5079235359 @default.
• W2803154237 hasAuthorship W2803154237A5088359990 @default.
• W2803154237 hasConcept C103278499 @default.
• W2803154237 hasConcept C104317684 @default.
• W2803154237 hasConcept C108583219 @default.
• W2803154237 hasConcept C115961682 @default.
• W2803154237 hasConcept C119857082 @default.
• W2803154237 hasConcept C150899416 @default.
• W2803154237 hasConcept C154945302 @default.
• W2803154237 hasConcept C177264268 @default.
• W2803154237 hasConcept C185592680 @default.
• W2803154237 hasConcept C199360897 @default.
• W2803154237 hasConcept C204241405 @default.
• W2803154237 hasConcept C22019652 @default.
• W2803154237 hasConcept C2776145971 @default.
• W2803154237 hasConcept C2984842247 @default.
• W2803154237 hasConcept C41008148 @default.
• W2803154237 hasConcept C50644808 @default.
• W2803154237 hasConcept C51632099 @default.
• W2803154237 hasConcept C55493867 @default.
• W2803154237 hasConceptScore W2803154237C103278499 @default.
• W2803154237 hasConceptScore W2803154237C104317684 @default.
• W2803154237 hasConceptScore W2803154237C108583219 @default.
• W2803154237 hasConceptScore W2803154237C115961682 @default.
• W2803154237 hasConceptScore W2803154237C119857082 @default.
• W2803154237 hasConceptScore W2803154237C150899416 @default.
• W2803154237 hasConceptScore W2803154237C154945302 @default.
• W2803154237 hasConceptScore W2803154237C177264268 @default.
• W2803154237 hasConceptScore W2803154237C185592680 @default.
• W2803154237 hasConceptScore W2803154237C199360897 @default.
• W2803154237 hasConceptScore W2803154237C204241405 @default.
• W2803154237 hasConceptScore W2803154237C22019652 @default.
• W2803154237 hasConceptScore W2803154237C2776145971 @default.
• W2803154237 hasConceptScore W2803154237C2984842247 @default.
• W2803154237 hasConceptScore W2803154237C41008148 @default.
• W2803154237 hasConceptScore W2803154237C50644808 @default.
• W2803154237 hasConceptScore W2803154237C51632099 @default.
• W2803154237 hasConceptScore W2803154237C55493867 @default.
• W2803154237 hasLocation W28031542371 @default.
• W2803154237 hasOpenAccess W2803154237 @default.
• W2803154237 hasPrimaryLocation W28031542371 @default.
• W2803154237 hasRelatedWork W1497443639 @default.
• W2803154237 hasRelatedWork W1542934499 @default.
• W2803154237 hasRelatedWork W1585902826 @default.
• W2803154237 hasRelatedWork W2051863572 @default.
• W2803154237 hasRelatedWork W2156726056 @default.
• W2803154237 hasRelatedWork W2194775991 @default.
• W2803154237 hasRelatedWork W2496665265 @default.
• W2803154237 hasRelatedWork W2731646070 @default.
• W2803154237 hasRelatedWork W2741312414 @default.
• W2803154237 hasRelatedWork W2903787679 @default.
• W2803154237 hasRelatedWork W2921006195 @default.
• W2803154237 hasRelatedWork W2936604471 @default.
• W2803154237 hasRelatedWork W2947188859 @default.

• next