Matches in SemOpenAlex for { <https://semopenalex.org/work/W3104108945> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W3104108945 endingPage "455" @default.
- W3104108945 startingPage "440" @default.
- W3104108945 abstract "In this paper, we present a deep neural network (DNN) training approach called the DeepMimic training method. Enormous amounts of data are available nowadays for training usage. Yet, only a tiny portion of these data is manually labeled, whereas almost all of the data are unlabeled. The training approach presented utilizes, in a most simplified manner, the unlabeled data to the fullest, in order to achieve remarkable (classification) results. Our DeepMimic method uses a small portion of labeled data and a large amount of unlabeled data for the training process, as expected in a real-world scenario. It consists of a mentor model and a student model. Employing a mentor model trained on a small portion of the labeled data and then feeding it only with unlabeled data, we show how to obtain a (simplified) student model that reaches the same accuracy and loss as the mentor model, on the same test set, without using any of the original data labels in the training of the student model. Our experiments demonstrate that even on challenging classification tasks the student network architecture can be simplified significantly with a minor influence on the performance, i.e., we need not even know the original network architecture of the mentor. In addition, the time required for training the student model to reach the mentor's performance level is shorter, as a result of a simplified architecture and more available data. The proposed method highlights the disadvantages of regular supervised training and demonstrates the benefits of a less traditional training approach." @default.
- W3104108945 created "2020-11-23" @default.
- W3104108945 creator A5018086675 @default.
- W3104108945 creator A5024299231 @default.
- W3104108945 creator A5025274307 @default.
- W3104108945 date "2019-01-01" @default.
- W3104108945 modified "2023-10-16" @default.
- W3104108945 title "DeepMimic: Mentor-Student Unlabeled Data Based Training" @default.
- W3104108945 cites W1902237438 @default.
- W3104108945 cites W2004915807 @default.
- W3104108945 cites W2108598243 @default.
- W3104108945 cites W2112796928 @default.
- W3104108945 cites W2194775991 @default.
- W3104108945 cites W2294370754 @default.
- W3104108945 cites W2752782242 @default.
- W3104108945 cites W2766447205 @default.
- W3104108945 cites W2796346823 @default.
- W3104108945 cites W2808195004 @default.
- W3104108945 cites W2903707108 @default.
- W3104108945 cites W2962792802 @default.
- W3104108945 cites W2963726581 @default.
- W3104108945 cites W2964081807 @default.
- W3104108945 cites W2968868585 @default.
- W3104108945 cites W3098637454 @default.
- W3104108945 cites W3106038055 @default.
- W3104108945 doi "https://doi.org/10.1007/978-3-030-30493-5_44" @default.
- W3104108945 hasPublicationYear "2019" @default.
- W3104108945 type Work @default.
- W3104108945 sameAs 3104108945 @default.
- W3104108945 citedByCount "2" @default.
- W3104108945 countsByYear W31041089452021 @default.
- W3104108945 countsByYear W31041089452023 @default.
- W3104108945 crossrefType "book-chapter" @default.
- W3104108945 hasAuthorship W3104108945A5018086675 @default.
- W3104108945 hasAuthorship W3104108945A5024299231 @default.
- W3104108945 hasAuthorship W3104108945A5025274307 @default.
- W3104108945 hasBestOaLocation W31041089452 @default.
- W3104108945 hasConcept C111919701 @default.
- W3104108945 hasConcept C115903868 @default.
- W3104108945 hasConcept C119857082 @default.
- W3104108945 hasConcept C121332964 @default.
- W3104108945 hasConcept C123657996 @default.
- W3104108945 hasConcept C124101348 @default.
- W3104108945 hasConcept C142362112 @default.
- W3104108945 hasConcept C153180895 @default.
- W3104108945 hasConcept C153294291 @default.
- W3104108945 hasConcept C153349607 @default.
- W3104108945 hasConcept C154945302 @default.
- W3104108945 hasConcept C16910744 @default.
- W3104108945 hasConcept C177264268 @default.
- W3104108945 hasConcept C199360897 @default.
- W3104108945 hasConcept C2776145971 @default.
- W3104108945 hasConcept C2777211547 @default.
- W3104108945 hasConcept C41008148 @default.
- W3104108945 hasConcept C50644808 @default.
- W3104108945 hasConcept C51632099 @default.
- W3104108945 hasConcept C58489278 @default.
- W3104108945 hasConcept C98045186 @default.
- W3104108945 hasConceptScore W3104108945C111919701 @default.
- W3104108945 hasConceptScore W3104108945C115903868 @default.
- W3104108945 hasConceptScore W3104108945C119857082 @default.
- W3104108945 hasConceptScore W3104108945C121332964 @default.
- W3104108945 hasConceptScore W3104108945C123657996 @default.
- W3104108945 hasConceptScore W3104108945C124101348 @default.
- W3104108945 hasConceptScore W3104108945C142362112 @default.
- W3104108945 hasConceptScore W3104108945C153180895 @default.
- W3104108945 hasConceptScore W3104108945C153294291 @default.
- W3104108945 hasConceptScore W3104108945C153349607 @default.
- W3104108945 hasConceptScore W3104108945C154945302 @default.
- W3104108945 hasConceptScore W3104108945C16910744 @default.
- W3104108945 hasConceptScore W3104108945C177264268 @default.
- W3104108945 hasConceptScore W3104108945C199360897 @default.
- W3104108945 hasConceptScore W3104108945C2776145971 @default.
- W3104108945 hasConceptScore W3104108945C2777211547 @default.
- W3104108945 hasConceptScore W3104108945C41008148 @default.
- W3104108945 hasConceptScore W3104108945C50644808 @default.
- W3104108945 hasConceptScore W3104108945C51632099 @default.
- W3104108945 hasConceptScore W3104108945C58489278 @default.
- W3104108945 hasConceptScore W3104108945C98045186 @default.
- W3104108945 hasLocation W31041089451 @default.
- W3104108945 hasLocation W31041089452 @default.
- W3104108945 hasOpenAccess W3104108945 @default.
- W3104108945 hasPrimaryLocation W31041089451 @default.
- W3104108945 hasRelatedWork W2012689841 @default.
- W3104108945 hasRelatedWork W2071508803 @default.
- W3104108945 hasRelatedWork W2122552724 @default.
- W3104108945 hasRelatedWork W2350386639 @default.
- W3104108945 hasRelatedWork W2360523936 @default.
- W3104108945 hasRelatedWork W2762452805 @default.
- W3104108945 hasRelatedWork W2772521755 @default.
- W3104108945 hasRelatedWork W2792951589 @default.
- W3104108945 hasRelatedWork W3104570445 @default.
- W3104108945 hasRelatedWork W4287595784 @default.
- W3104108945 isParatext "false" @default.
- W3104108945 isRetracted "false" @default.
- W3104108945 magId "3104108945" @default.
- W3104108945 workType "book-chapter" @default.