Matches in SemOpenAlex for { <https://semopenalex.org/work/W4300643609> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4300643609 abstract "Visual recognition algorithms are required today to exhibit adaptive abilities. Given a deep model trained on a specific, given task, it would be highly desirable to be able to adapt incrementally to new tasks, preserving scalability as the number of new tasks increases, while at the same time avoiding catastrophic forgetting issues. Recent work has shown that masking the internal weights of a given original conv-net through learned binary variables is a promising strategy. We build upon this intuition and take into account more elaborated affine transformations of the convolutional weights that include learned binary masks. We show that with our generalization it is possible to achieve significantly higher levels of adaptation to new tasks, enabling the approach to compete with fine tuning strategies by requiring slightly more than 1 bit per network parameter per additional task. Experiments on two popular benchmarks showcase the power of our approach, that achieves the new state of the art on the Visual Decathlon Challenge." @default.
- W4300643609 created "2022-10-03" @default.
- W4300643609 creator A5017971549 @default.
- W4300643609 creator A5024757596 @default.
- W4300643609 creator A5065059558 @default.
- W4300643609 creator A5073916921 @default.
- W4300643609 date "2018-05-28" @default.
- W4300643609 modified "2023-10-17" @default.
- W4300643609 title "Adding New Tasks to a Single Network with Weight Transformations using Binary Masks" @default.
- W4300643609 doi "https://doi.org/10.48550/arxiv.1805.11119" @default.
- W4300643609 hasPublicationYear "2018" @default.
- W4300643609 type Work @default.
- W4300643609 citedByCount "0" @default.
- W4300643609 crossrefType "posted-content" @default.
- W4300643609 hasAuthorship W4300643609A5017971549 @default.
- W4300643609 hasAuthorship W4300643609A5024757596 @default.
- W4300643609 hasAuthorship W4300643609A5065059558 @default.
- W4300643609 hasAuthorship W4300643609A5073916921 @default.
- W4300643609 hasBestOaLocation W43006436091 @default.
- W4300643609 hasConcept C119857082 @default.
- W4300643609 hasConcept C134306372 @default.
- W4300643609 hasConcept C138885662 @default.
- W4300643609 hasConcept C154945302 @default.
- W4300643609 hasConcept C162324750 @default.
- W4300643609 hasConcept C177148314 @default.
- W4300643609 hasConcept C187736073 @default.
- W4300643609 hasConcept C202444582 @default.
- W4300643609 hasConcept C2780451532 @default.
- W4300643609 hasConcept C33923547 @default.
- W4300643609 hasConcept C41008148 @default.
- W4300643609 hasConcept C41895202 @default.
- W4300643609 hasConcept C48044578 @default.
- W4300643609 hasConcept C48372109 @default.
- W4300643609 hasConcept C7149132 @default.
- W4300643609 hasConcept C77088390 @default.
- W4300643609 hasConcept C80444323 @default.
- W4300643609 hasConcept C92757383 @default.
- W4300643609 hasConcept C94375191 @default.
- W4300643609 hasConceptScore W4300643609C119857082 @default.
- W4300643609 hasConceptScore W4300643609C134306372 @default.
- W4300643609 hasConceptScore W4300643609C138885662 @default.
- W4300643609 hasConceptScore W4300643609C154945302 @default.
- W4300643609 hasConceptScore W4300643609C162324750 @default.
- W4300643609 hasConceptScore W4300643609C177148314 @default.
- W4300643609 hasConceptScore W4300643609C187736073 @default.
- W4300643609 hasConceptScore W4300643609C202444582 @default.
- W4300643609 hasConceptScore W4300643609C2780451532 @default.
- W4300643609 hasConceptScore W4300643609C33923547 @default.
- W4300643609 hasConceptScore W4300643609C41008148 @default.
- W4300643609 hasConceptScore W4300643609C41895202 @default.
- W4300643609 hasConceptScore W4300643609C48044578 @default.
- W4300643609 hasConceptScore W4300643609C48372109 @default.
- W4300643609 hasConceptScore W4300643609C7149132 @default.
- W4300643609 hasConceptScore W4300643609C77088390 @default.
- W4300643609 hasConceptScore W4300643609C80444323 @default.
- W4300643609 hasConceptScore W4300643609C92757383 @default.
- W4300643609 hasConceptScore W4300643609C94375191 @default.
- W4300643609 hasLocation W43006436091 @default.
- W4300643609 hasOpenAccess W4300643609 @default.
- W4300643609 hasPrimaryLocation W43006436091 @default.
- W4300643609 hasRelatedWork W1525643724 @default.
- W4300643609 hasRelatedWork W1964624709 @default.
- W4300643609 hasRelatedWork W2302028273 @default.
- W4300643609 hasRelatedWork W2364921833 @default.
- W4300643609 hasRelatedWork W2388030554 @default.
- W4300643609 hasRelatedWork W2461970972 @default.
- W4300643609 hasRelatedWork W3021857098 @default.
- W4300643609 hasRelatedWork W3204153069 @default.
- W4300643609 hasRelatedWork W4286954311 @default.
- W4300643609 hasRelatedWork W4320853675 @default.
- W4300643609 isParatext "false" @default.
- W4300643609 isRetracted "false" @default.
- W4300643609 workType "article" @default.