Matches in SemOpenAlex for { <https://semopenalex.org/work/W4286231696> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4286231696 abstract "Neural networks have seen an explosion of usage and research in the past decade, particularly within the domains of computer vision and natural language processing. However, only recently have advancements in neural networks yielded performance improvements beyond narrow applications and translated to expanded multitask models capable of generalizing across multiple data types and modalities. Simultaneously, it has been shown that neural networks are overparameterized to a high degree, and pruning techniques have proved capable of significantly reducing the number of active weights within the network while largely preserving performance. In this work, we identify a methodology and network representational structure which allows a pruned network to employ previously unused weights to learn subsequent tasks. We employ these methodologies on well-known benchmarking datasets for testing purposes and show that networks trained using our approaches are able to learn multiple tasks, which may be related or unrelated, in parallel or in sequence without sacrificing performance on any task or exhibiting catastrophic forgetting." @default.
- W4286231696 created "2022-07-21" @default.
- W4286231696 creator A5017433007 @default.
- W4286231696 creator A5044021585 @default.
- W4286231696 creator A5047107853 @default.
- W4286231696 creator A5086747498 @default.
- W4286231696 date "2022-07-18" @default.
- W4286231696 modified "2023-10-14" @default.
- W4286231696 title "The Multiple Subnetwork Hypothesis: Enabling Multidomain Learning by Isolating Task-Specific Subnetworks in Feedforward Neural Networks" @default.
- W4286231696 doi "https://doi.org/10.48550/arxiv.2207.08821" @default.
- W4286231696 hasPublicationYear "2022" @default.
- W4286231696 type Work @default.
- W4286231696 citedByCount "0" @default.
- W4286231696 crossrefType "posted-content" @default.
- W4286231696 hasAuthorship W4286231696A5017433007 @default.
- W4286231696 hasAuthorship W4286231696A5044021585 @default.
- W4286231696 hasAuthorship W4286231696A5047107853 @default.
- W4286231696 hasAuthorship W4286231696A5086747498 @default.
- W4286231696 hasBestOaLocation W42862316961 @default.
- W4286231696 hasConcept C108010975 @default.
- W4286231696 hasConcept C119857082 @default.
- W4286231696 hasConcept C127413603 @default.
- W4286231696 hasConcept C133731056 @default.
- W4286231696 hasConcept C138885662 @default.
- W4286231696 hasConcept C144024400 @default.
- W4286231696 hasConcept C144133560 @default.
- W4286231696 hasConcept C154945302 @default.
- W4286231696 hasConcept C162324750 @default.
- W4286231696 hasConcept C162853370 @default.
- W4286231696 hasConcept C187736073 @default.
- W4286231696 hasConcept C2779903281 @default.
- W4286231696 hasConcept C2780186347 @default.
- W4286231696 hasConcept C2780451532 @default.
- W4286231696 hasConcept C36289849 @default.
- W4286231696 hasConcept C38652104 @default.
- W4286231696 hasConcept C38858127 @default.
- W4286231696 hasConcept C41008148 @default.
- W4286231696 hasConcept C41895202 @default.
- W4286231696 hasConcept C50644808 @default.
- W4286231696 hasConcept C6557445 @default.
- W4286231696 hasConcept C7149132 @default.
- W4286231696 hasConcept C86251818 @default.
- W4286231696 hasConcept C86803240 @default.
- W4286231696 hasConceptScore W4286231696C108010975 @default.
- W4286231696 hasConceptScore W4286231696C119857082 @default.
- W4286231696 hasConceptScore W4286231696C127413603 @default.
- W4286231696 hasConceptScore W4286231696C133731056 @default.
- W4286231696 hasConceptScore W4286231696C138885662 @default.
- W4286231696 hasConceptScore W4286231696C144024400 @default.
- W4286231696 hasConceptScore W4286231696C144133560 @default.
- W4286231696 hasConceptScore W4286231696C154945302 @default.
- W4286231696 hasConceptScore W4286231696C162324750 @default.
- W4286231696 hasConceptScore W4286231696C162853370 @default.
- W4286231696 hasConceptScore W4286231696C187736073 @default.
- W4286231696 hasConceptScore W4286231696C2779903281 @default.
- W4286231696 hasConceptScore W4286231696C2780186347 @default.
- W4286231696 hasConceptScore W4286231696C2780451532 @default.
- W4286231696 hasConceptScore W4286231696C36289849 @default.
- W4286231696 hasConceptScore W4286231696C38652104 @default.
- W4286231696 hasConceptScore W4286231696C38858127 @default.
- W4286231696 hasConceptScore W4286231696C41008148 @default.
- W4286231696 hasConceptScore W4286231696C41895202 @default.
- W4286231696 hasConceptScore W4286231696C50644808 @default.
- W4286231696 hasConceptScore W4286231696C6557445 @default.
- W4286231696 hasConceptScore W4286231696C7149132 @default.
- W4286231696 hasConceptScore W4286231696C86251818 @default.
- W4286231696 hasConceptScore W4286231696C86803240 @default.
- W4286231696 hasLocation W42862316961 @default.
- W4286231696 hasOpenAccess W4286231696 @default.
- W4286231696 hasPrimaryLocation W42862316961 @default.
- W4286231696 hasRelatedWork W10961682 @default.
- W4286231696 hasRelatedWork W13538861 @default.
- W4286231696 hasRelatedWork W1383942 @default.
- W4286231696 hasRelatedWork W1407330 @default.
- W4286231696 hasRelatedWork W14516383 @default.
- W4286231696 hasRelatedWork W2683128 @default.
- W4286231696 hasRelatedWork W5983547 @default.
- W4286231696 hasRelatedWork W6661183 @default.
- W4286231696 hasRelatedWork W67111 @default.
- W4286231696 hasRelatedWork W8595301 @default.
- W4286231696 isParatext "false" @default.
- W4286231696 isRetracted "false" @default.
- W4286231696 workType "article" @default.