FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384302776> ?p ?o ?g. }

• W4384302776 abstract "Training deep neural network (DNN) models, which has become an important task in today's software development, is often costly in terms of computational resources and time. With the inspiration of software reuse, building DNN models through reusing existing ones has gained increasing attention recently. Prior approaches to DNN model reuse have two main limitations: 1) reusing the entire model, while only a small part of the model's functionalities (labels) are required, would cause much overhead (e.g., computational and time costs for inference), and 2) model reuse would inherit the defects and weaknesses of the reused model, and hence put the new system under threats of security attack. To solve the above problem, we propose SeaM, a tool that re-engineers a trained DNN model to improve its reusability. Specifically, given a target problem and a trained model, SeaM utilizes a gradient-based search method to search for the model's weights that are relevant to the target problem. The re-engineered model that only retains the relevant weights is then reused to solve the target problem. Evaluation results on widely-used models show that the re-engineered models produced by SeaM only contain 10.11% weights of the original models, resulting 42.41% reduction in terms of inference time. For the target problem, the re-engineered models even outperform the original models in classification accuracy by 5.85%. Moreover, reusing the re-engineered models inherits an average of 57% fewer defects than reusing the entire model. We believe our approach to reducing reuse overhead and defect inheritance is one important step forward for practical model reuse." @default.
• W4384302776 created "2023-07-15" @default.
• W4384302776 creator A5017712129 @default.
• W4384302776 creator A5022746945 @default.
• W4384302776 creator A5023350285 @default.
• W4384302776 creator A5028983129 @default.
• W4384302776 creator A5058691896 @default.
• W4384302776 creator A5069172429 @default.
• W4384302776 date "2023-05-01" @default.
• W4384302776 modified "2023-10-17" @default.
• W4384302776 title "Reusing Deep Neural Network Models through Model Re-engineering" @default.
• W4384302776 cites W1954152232 @default.
• W4384302776 cites W2006267758 @default.
• W4384302776 cites W2038765747 @default.
• W4384302776 cites W2095577883 @default.
• W4384302776 cites W2102605133 @default.
• W4384302776 cites W2108598243 @default.
• W4384302776 cites W2152161678 @default.
• W4384302776 cites W2165698076 @default.
• W4384302776 cites W2194775991 @default.
• W4384302776 cites W2285660444 @default.
• W4384302776 cites W2294861651 @default.
• W4384302776 cites W2533598788 @default.
• W4384302776 cites W2616028256 @default.
• W4384302776 cites W2804935296 @default.
• W4384302776 cites W2809254203 @default.
• W4384302776 cites W2888824816 @default.
• W4384302776 cites W2897865027 @default.
• W4384302776 cites W2954678575 @default.
• W4384302776 cites W2963460174 @default.
• W4384302776 cites W2964233199 @default.
• W4384302776 cites W2994987245 @default.
• W4384302776 cites W3008515144 @default.
• W4384302776 cites W3082665562 @default.
• W4384302776 cites W3103697033 @default.
• W4384302776 cites W3163829544 @default.
• W4384302776 cites W3169839597 @default.
• W4384302776 cites W3176064809 @default.
• W4384302776 cites W3207416911 @default.
• W4384302776 cites W4284669642 @default.
• W4384302776 cites W4313563686 @default.
• W4384302776 cites W2981431987 @default.
• W4384302776 doi "https://doi.org/10.1109/icse48619.2023.00090" @default.
• W4384302776 hasPublicationYear "2023" @default.
• W4384302776 type Work @default.
• W4384302776 citedByCount "0" @default.
• W4384302776 crossrefType "proceedings-article" @default.
• W4384302776 hasAuthorship W4384302776A5017712129 @default.
• W4384302776 hasAuthorship W4384302776A5022746945 @default.
• W4384302776 hasAuthorship W4384302776A5023350285 @default.
• W4384302776 hasAuthorship W4384302776A5028983129 @default.
• W4384302776 hasAuthorship W4384302776A5058691896 @default.
• W4384302776 hasAuthorship W4384302776A5069172429 @default.
• W4384302776 hasBestOaLocation W43843027762 @default.
• W4384302776 hasConcept C111335779 @default.
• W4384302776 hasConcept C115903868 @default.
• W4384302776 hasConcept C119857082 @default.
• W4384302776 hasConcept C127413603 @default.
• W4384302776 hasConcept C137981799 @default.
• W4384302776 hasConcept C154945302 @default.
• W4384302776 hasConcept C199360897 @default.
• W4384302776 hasConcept C201995342 @default.
• W4384302776 hasConcept C206588197 @default.
• W4384302776 hasConcept C2524010 @default.
• W4384302776 hasConcept C2776214188 @default.
• W4384302776 hasConcept C2777904410 @default.
• W4384302776 hasConcept C2779960059 @default.
• W4384302776 hasConcept C2780451532 @default.
• W4384302776 hasConcept C33923547 @default.
• W4384302776 hasConcept C41008148 @default.
• W4384302776 hasConcept C50644808 @default.
• W4384302776 hasConcept C548081761 @default.
• W4384302776 hasConceptScore W4384302776C111335779 @default.
• W4384302776 hasConceptScore W4384302776C115903868 @default.
• W4384302776 hasConceptScore W4384302776C119857082 @default.
• W4384302776 hasConceptScore W4384302776C127413603 @default.
• W4384302776 hasConceptScore W4384302776C137981799 @default.
• W4384302776 hasConceptScore W4384302776C154945302 @default.
• W4384302776 hasConceptScore W4384302776C199360897 @default.
• W4384302776 hasConceptScore W4384302776C201995342 @default.
• W4384302776 hasConceptScore W4384302776C206588197 @default.
• W4384302776 hasConceptScore W4384302776C2524010 @default.
• W4384302776 hasConceptScore W4384302776C2776214188 @default.
• W4384302776 hasConceptScore W4384302776C2777904410 @default.
• W4384302776 hasConceptScore W4384302776C2779960059 @default.
• W4384302776 hasConceptScore W4384302776C2780451532 @default.
• W4384302776 hasConceptScore W4384302776C33923547 @default.
• W4384302776 hasConceptScore W4384302776C41008148 @default.
• W4384302776 hasConceptScore W4384302776C50644808 @default.
• W4384302776 hasConceptScore W4384302776C548081761 @default.
• W4384302776 hasFunder F4320321001 @default.
• W4384302776 hasFunder F4320334704 @default.
• W4384302776 hasLocation W43843027761 @default.
• W4384302776 hasLocation W43843027762 @default.
• W4384302776 hasOpenAccess W4384302776 @default.
• W4384302776 hasPrimaryLocation W43843027761 @default.
• W4384302776 hasRelatedWork W1791926134 @default.
• W4384302776 hasRelatedWork W2111589395 @default.
• W4384302776 hasRelatedWork W2157615579 @default.
• W4384302776 hasRelatedWork W2350449555 @default.

• next