FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3118781314> ?p ?o ?g. }

• W3118781314 abstract "Batch normalization plays a crucial role when training deep neural networks. However, batch statistics become unstable with small batch sizes and are unreliable in the presence of distribution shifts. We propose MetaNorm, a simple yet effective meta-learning normalization. It tackles the aforementioned issues in a unified way by leveraging the meta-learning setting and learns to infer adaptive statistics for batch normalization. MetaNorm is generic, flexible and model-agnostic, making it a simple plug-and-play module that is seamlessly embedded into existing meta-learning approaches. It can be efficiently implemented by lightweight hypernetworks with low computational cost. We verify its effectiveness by extensive evaluation on representative tasks suffering from the small batch and domain shift problems: few-shot learning and domain generalization. We further introduce an even more challenging setting: few-shot domain generalization. Results demonstrate that MetaNorm consistently achieves better, or at least competitive, accuracy compared to existing batch normalization methods." @default.
• W3118781314 created "2021-01-18" @default.
• W3118781314 creator A5024508073 @default.
• W3118781314 creator A5026486701 @default.
• W3118781314 creator A5077690631 @default.
• W3118781314 creator A5082634513 @default.
• W3118781314 date "2021-05-03" @default.
• W3118781314 modified "2023-09-23" @default.
• W3118781314 title "MetaNorm: Learning to Normalize Few-Shot Batches Across Domains" @default.
• W3118781314 cites W1797268635 @default.
• W3118781314 cites W1836465849 @default.
• W3118781314 cites W1846799578 @default.
• W3118781314 cites W1861492603 @default.
• W3118781314 cites W1959608418 @default.
• W3118781314 cites W2002427601 @default.
• W3118781314 cites W2047643928 @default.
• W3118781314 cites W2117539524 @default.
• W3118781314 cites W2155858138 @default.
• W3118781314 cites W2194321275 @default.
• W3118781314 cites W2299668505 @default.
• W3118781314 cites W2502312327 @default.
• W3118781314 cites W2533598788 @default.
• W3118781314 cites W2581955877 @default.
• W3118781314 cites W2601450892 @default.
• W3118781314 cites W2604763608 @default.
• W3118781314 cites W2627183927 @default.
• W3118781314 cites W2753160622 @default.
• W3118781314 cites W2763549966 @default.
• W3118781314 cites W2795900505 @default.
• W3118781314 cites W2798658180 @default.
• W3118781314 cites W2884366600 @default.
• W3118781314 cites W2889965839 @default.
• W3118781314 cites W2912528108 @default.
• W3118781314 cites W2927966956 @default.
• W3118781314 cites W2949813473 @default.
• W3118781314 cites W2952258850 @default.
• W3118781314 cites W2958360136 @default.
• W3118781314 cites W2962723986 @default.
• W3118781314 cites W2963245493 @default.
• W3118781314 cites W2963300719 @default.
• W3118781314 cites W2963341924 @default.
• W3118781314 cites W2963371846 @default.
• W3118781314 cites W2964068654 @default.
• W3118781314 cites W2970987681 @default.
• W3118781314 cites W2981720610 @default.
• W3118781314 cites W2996598013 @default.
• W3118781314 cites W2996623013 @default.
• W3118781314 cites W3004880962 @default.
• W3118781314 cites W3035543253 @default.
• W3118781314 cites W3036825277 @default.
• W3118781314 cites W3104140305 @default.
• W3118781314 cites W3109042658 @default.
• W3118781314 cites W3109831211 @default.
• W3118781314 cites W3110608229 @default.
• W3118781314 hasPublicationYear "2021" @default.
• W3118781314 type Work @default.
• W3118781314 sameAs 3118781314 @default.
• W3118781314 citedByCount "7" @default.
• W3118781314 countsByYear W31187813142021 @default.
• W3118781314 countsByYear W31187813142022 @default.
• W3118781314 countsByYear W31187813142023 @default.
• W3118781314 crossrefType "proceedings-article" @default.
• W3118781314 hasAuthorship W3118781314A5024508073 @default.
• W3118781314 hasAuthorship W3118781314A5026486701 @default.
• W3118781314 hasAuthorship W3118781314A5077690631 @default.
• W3118781314 hasAuthorship W3118781314A5082634513 @default.
• W3118781314 hasConcept C108583219 @default.
• W3118781314 hasConcept C119857082 @default.
• W3118781314 hasConcept C134306372 @default.
• W3118781314 hasConcept C136886441 @default.
• W3118781314 hasConcept C144024400 @default.
• W3118781314 hasConcept C154945302 @default.
• W3118781314 hasConcept C162324750 @default.
• W3118781314 hasConcept C177148314 @default.
• W3118781314 hasConcept C187736073 @default.
• W3118781314 hasConcept C19165224 @default.
• W3118781314 hasConcept C2780451532 @default.
• W3118781314 hasConcept C2781002164 @default.
• W3118781314 hasConcept C2984842247 @default.
• W3118781314 hasConcept C33923547 @default.
• W3118781314 hasConcept C41008148 @default.
• W3118781314 hasConcept C50644808 @default.
• W3118781314 hasConceptScore W3118781314C108583219 @default.
• W3118781314 hasConceptScore W3118781314C119857082 @default.
• W3118781314 hasConceptScore W3118781314C134306372 @default.
• W3118781314 hasConceptScore W3118781314C136886441 @default.
• W3118781314 hasConceptScore W3118781314C144024400 @default.
• W3118781314 hasConceptScore W3118781314C154945302 @default.
• W3118781314 hasConceptScore W3118781314C162324750 @default.
• W3118781314 hasConceptScore W3118781314C177148314 @default.
• W3118781314 hasConceptScore W3118781314C187736073 @default.
• W3118781314 hasConceptScore W3118781314C19165224 @default.
• W3118781314 hasConceptScore W3118781314C2780451532 @default.
• W3118781314 hasConceptScore W3118781314C2781002164 @default.
• W3118781314 hasConceptScore W3118781314C2984842247 @default.
• W3118781314 hasConceptScore W3118781314C33923547 @default.
• W3118781314 hasConceptScore W3118781314C41008148 @default.
• W3118781314 hasConceptScore W3118781314C50644808 @default.
• W3118781314 hasOpenAccess W3118781314 @default.
• W3118781314 hasRelatedWork W2194775991 @default.

• next