Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366343221> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4366343221 abstract "Non-exemplar class-incremental learning refers to classifying new and old classes without storing samples of old classes. Since only new class samples are available for optimization, it often occurs catastrophic forgetting of old knowledge. To alleviate this problem, many new methods are proposed such as model distillation, class augmentation. In this paper, we propose an effective non-exemplar method called RAMF consisting of Random Auxiliary classes augmentation and Mixed Feature. On the one hand, we design a novel random auxiliary classes augmentation method, where one augmentation is randomly selected from three augmentations and applied on the input to generate augmented samples and extra class labels. By extending data and label space, it allows the model to learn more diverse representations, which can prevent the model from being biased towards learning task-specific features. When learning new tasks, it will reduce the change of feature space and improve model generalization. On the other hand, we employ mixed feature to replace the new features since only using new feature to optimize the model will affect the representation that was previously embedded in the feature space. Instead, by mixing new and old features, old knowledge can be retained without increasing the computational complexity. Extensive experiments on three benchmarks demonstrate the superiority of our approach, which outperforms the state-of-the-art non-exemplar methods and is comparable to high-performance replay-based methods." @default.
- W4366343221 created "2023-04-20" @default.
- W4366343221 creator A5008081224 @default.
- W4366343221 creator A5011496755 @default.
- W4366343221 creator A5028235866 @default.
- W4366343221 creator A5043715429 @default.
- W4366343221 creator A5050711402 @default.
- W4366343221 date "2023-04-16" @default.
- W4366343221 modified "2023-10-12" @default.
- W4366343221 title "Non-exemplar Class-incremental Learning by Random Auxiliary Classes Augmentation and Mixed Features" @default.
- W4366343221 doi "https://doi.org/10.48550/arxiv.2304.07707" @default.
- W4366343221 hasPublicationYear "2023" @default.
- W4366343221 type Work @default.
- W4366343221 citedByCount "0" @default.
- W4366343221 crossrefType "posted-content" @default.
- W4366343221 hasAuthorship W4366343221A5008081224 @default.
- W4366343221 hasAuthorship W4366343221A5011496755 @default.
- W4366343221 hasAuthorship W4366343221A5028235866 @default.
- W4366343221 hasAuthorship W4366343221A5043715429 @default.
- W4366343221 hasAuthorship W4366343221A5050711402 @default.
- W4366343221 hasBestOaLocation W43663432211 @default.
- W4366343221 hasConcept C111919701 @default.
- W4366343221 hasConcept C119857082 @default.
- W4366343221 hasConcept C134306372 @default.
- W4366343221 hasConcept C138885662 @default.
- W4366343221 hasConcept C153180895 @default.
- W4366343221 hasConcept C154945302 @default.
- W4366343221 hasConcept C162324750 @default.
- W4366343221 hasConcept C177148314 @default.
- W4366343221 hasConcept C17744445 @default.
- W4366343221 hasConcept C187736073 @default.
- W4366343221 hasConcept C199539241 @default.
- W4366343221 hasConcept C2776359362 @default.
- W4366343221 hasConcept C2776401178 @default.
- W4366343221 hasConcept C2777212361 @default.
- W4366343221 hasConcept C2778572836 @default.
- W4366343221 hasConcept C2780451532 @default.
- W4366343221 hasConcept C33923547 @default.
- W4366343221 hasConcept C41008148 @default.
- W4366343221 hasConcept C41895202 @default.
- W4366343221 hasConcept C7149132 @default.
- W4366343221 hasConcept C83665646 @default.
- W4366343221 hasConcept C94625758 @default.
- W4366343221 hasConceptScore W4366343221C111919701 @default.
- W4366343221 hasConceptScore W4366343221C119857082 @default.
- W4366343221 hasConceptScore W4366343221C134306372 @default.
- W4366343221 hasConceptScore W4366343221C138885662 @default.
- W4366343221 hasConceptScore W4366343221C153180895 @default.
- W4366343221 hasConceptScore W4366343221C154945302 @default.
- W4366343221 hasConceptScore W4366343221C162324750 @default.
- W4366343221 hasConceptScore W4366343221C177148314 @default.
- W4366343221 hasConceptScore W4366343221C17744445 @default.
- W4366343221 hasConceptScore W4366343221C187736073 @default.
- W4366343221 hasConceptScore W4366343221C199539241 @default.
- W4366343221 hasConceptScore W4366343221C2776359362 @default.
- W4366343221 hasConceptScore W4366343221C2776401178 @default.
- W4366343221 hasConceptScore W4366343221C2777212361 @default.
- W4366343221 hasConceptScore W4366343221C2778572836 @default.
- W4366343221 hasConceptScore W4366343221C2780451532 @default.
- W4366343221 hasConceptScore W4366343221C33923547 @default.
- W4366343221 hasConceptScore W4366343221C41008148 @default.
- W4366343221 hasConceptScore W4366343221C41895202 @default.
- W4366343221 hasConceptScore W4366343221C7149132 @default.
- W4366343221 hasConceptScore W4366343221C83665646 @default.
- W4366343221 hasConceptScore W4366343221C94625758 @default.
- W4366343221 hasLocation W43663432211 @default.
- W4366343221 hasOpenAccess W4366343221 @default.
- W4366343221 hasPrimaryLocation W43663432211 @default.
- W4366343221 hasRelatedWork W1964624709 @default.
- W4366343221 hasRelatedWork W2096020108 @default.
- W4366343221 hasRelatedWork W2117297561 @default.
- W4366343221 hasRelatedWork W2177773744 @default.
- W4366343221 hasRelatedWork W2178080893 @default.
- W4366343221 hasRelatedWork W2546942002 @default.
- W4366343221 hasRelatedWork W2970216048 @default.
- W4366343221 hasRelatedWork W4281760909 @default.
- W4366343221 hasRelatedWork W4362598752 @default.
- W4366343221 hasRelatedWork W2178852694 @default.
- W4366343221 isParatext "false" @default.
- W4366343221 isRetracted "false" @default.
- W4366343221 workType "article" @default.