Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312479215> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W4312479215 endingPage "117010" @default.
- W4312479215 startingPage "117001" @default.
- W4312479215 abstract "We introduce a neural network architecture that logarithmically reduces the number of self-rehearsal steps in the generative rehearsal of continually learned models. In continual learning (CL), training samples come in subsequent tasks, and the trained model can access only a current task. Contemporary CL methods employ generative models to replay previous samples and train them recursively with a combination of current and regenerated past data. This recurrence leads to superfluous computations as the same past samples are regenerated after each task, and the reconstruction quality successively degrades. In this work, we address these limitations and propose a new generative rehearsal architecture that requires, at most, a logarithmic number of retraining sessions for each sample. Our approach leverages the allocation of past data in a set of generative models such that most of them do not require retraining after a task. The experimental evaluation of our logarithmic continual learning approach shows the superiority of our method with respect to the state-of-the-art generative rehearsal methods." @default.
- W4312479215 created "2023-01-05" @default.
- W4312479215 creator A5026378980 @default.
- W4312479215 creator A5056517464 @default.
- W4312479215 creator A5070627781 @default.
- W4312479215 creator A5078472796 @default.
- W4312479215 creator A5088799131 @default.
- W4312479215 date "2022-01-01" @default.
- W4312479215 modified "2023-09-26" @default.
- W4312479215 title "Logarithmic Continual Learning" @default.
- W4312479215 cites W2060277733 @default.
- W4312479215 cites W2194321275 @default.
- W4312479215 cites W2560647685 @default.
- W4312479215 cites W2787295326 @default.
- W4312479215 cites W2954929116 @default.
- W4312479215 cites W2964189064 @default.
- W4312479215 cites W2968596670 @default.
- W4312479215 cites W2982110666 @default.
- W4312479215 cites W3003159499 @default.
- W4312479215 cites W3042642865 @default.
- W4312479215 cites W3097784654 @default.
- W4312479215 cites W3125116114 @default.
- W4312479215 cites W3203936539 @default.
- W4312479215 doi "https://doi.org/10.1109/access.2022.3218907" @default.
- W4312479215 hasPublicationYear "2022" @default.
- W4312479215 type Work @default.
- W4312479215 citedByCount "0" @default.
- W4312479215 crossrefType "journal-article" @default.
- W4312479215 hasAuthorship W4312479215A5026378980 @default.
- W4312479215 hasAuthorship W4312479215A5056517464 @default.
- W4312479215 hasAuthorship W4312479215A5070627781 @default.
- W4312479215 hasAuthorship W4312479215A5078472796 @default.
- W4312479215 hasAuthorship W4312479215A5088799131 @default.
- W4312479215 hasBestOaLocation W43124792151 @default.
- W4312479215 hasConcept C119857082 @default.
- W4312479215 hasConcept C134306372 @default.
- W4312479215 hasConcept C144133560 @default.
- W4312479215 hasConcept C154945302 @default.
- W4312479215 hasConcept C155202549 @default.
- W4312479215 hasConcept C162324750 @default.
- W4312479215 hasConcept C167966045 @default.
- W4312479215 hasConcept C177264268 @default.
- W4312479215 hasConcept C185592680 @default.
- W4312479215 hasConcept C187736073 @default.
- W4312479215 hasConcept C198531522 @default.
- W4312479215 hasConcept C199360897 @default.
- W4312479215 hasConcept C2778712577 @default.
- W4312479215 hasConcept C2780451532 @default.
- W4312479215 hasConcept C33923547 @default.
- W4312479215 hasConcept C39890363 @default.
- W4312479215 hasConcept C39927690 @default.
- W4312479215 hasConcept C41008148 @default.
- W4312479215 hasConcept C43617362 @default.
- W4312479215 hasConcept C50644808 @default.
- W4312479215 hasConceptScore W4312479215C119857082 @default.
- W4312479215 hasConceptScore W4312479215C134306372 @default.
- W4312479215 hasConceptScore W4312479215C144133560 @default.
- W4312479215 hasConceptScore W4312479215C154945302 @default.
- W4312479215 hasConceptScore W4312479215C155202549 @default.
- W4312479215 hasConceptScore W4312479215C162324750 @default.
- W4312479215 hasConceptScore W4312479215C167966045 @default.
- W4312479215 hasConceptScore W4312479215C177264268 @default.
- W4312479215 hasConceptScore W4312479215C185592680 @default.
- W4312479215 hasConceptScore W4312479215C187736073 @default.
- W4312479215 hasConceptScore W4312479215C198531522 @default.
- W4312479215 hasConceptScore W4312479215C199360897 @default.
- W4312479215 hasConceptScore W4312479215C2778712577 @default.
- W4312479215 hasConceptScore W4312479215C2780451532 @default.
- W4312479215 hasConceptScore W4312479215C33923547 @default.
- W4312479215 hasConceptScore W4312479215C39890363 @default.
- W4312479215 hasConceptScore W4312479215C39927690 @default.
- W4312479215 hasConceptScore W4312479215C41008148 @default.
- W4312479215 hasConceptScore W4312479215C43617362 @default.
- W4312479215 hasConceptScore W4312479215C50644808 @default.
- W4312479215 hasFunder F4320321042 @default.
- W4312479215 hasFunder F4320322511 @default.
- W4312479215 hasLocation W43124792151 @default.
- W4312479215 hasLocation W43124792152 @default.
- W4312479215 hasOpenAccess W4312479215 @default.
- W4312479215 hasPrimaryLocation W43124792151 @default.
- W4312479215 hasRelatedWork W1534961803 @default.
- W4312479215 hasRelatedWork W2770426046 @default.
- W4312479215 hasRelatedWork W2874782909 @default.
- W4312479215 hasRelatedWork W2952072295 @default.
- W4312479215 hasRelatedWork W4226454691 @default.
- W4312479215 hasRelatedWork W4289760695 @default.
- W4312479215 hasRelatedWork W4299531873 @default.
- W4312479215 hasRelatedWork W4312479215 @default.
- W4312479215 hasRelatedWork W4385572368 @default.
- W4312479215 hasRelatedWork W2310403681 @default.
- W4312479215 hasVolume "10" @default.
- W4312479215 isParatext "false" @default.
- W4312479215 isRetracted "false" @default.
- W4312479215 workType "article" @default.