Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387319552> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4387319552 endingPage "209" @default.
- W4387319552 startingPage "196" @default.
- W4387319552 abstract "Machine learning (ML) and Deep learning (DL) have been widely studied and adopted for different applications across various fields. There is a growing demand for ML implementations as well as ML accelerators for small devices for Internet-of-Things (IoT) applications. Often, these accelerators allow efficient edge-based inference based on pre-trained deep neural network models for IoT setting. First, the model will be trained separately on a more powerful machine and then deployed on the edge device for inference. However, there are several attacks reported that could recover and steal the pre-trained model. For example, recently an attack was reported on edge-based machine learning accelerator demonstrated recovery of target neural network models (architecture and weights) using cold-boot attack. Using this information, the adversary can reconstruct the model, albeit with certain errors due to the corruption of the data during the recovery process. Hence, this indicate potential vulnerability of implementation of ML/DL model on edge devices for IoT applications. In this work, we investigate generic countermeasures for model recovery attacks, based on neural network (NN) model optimization technique, such as quantization, binarization, pruning, etc. We first study and investigate the performance improvement offered and how these transformations could help in mitigating the model recovery process. Our experimental results show that model optimization methods, in addition to achieving better performance, can result in accuracy degradation which help to mitigate model recovery attacks." @default.
- W4387319552 created "2023-10-04" @default.
- W4387319552 creator A5060395435 @default.
- W4387319552 creator A5081720773 @default.
- W4387319552 date "2023-01-01" @default.
- W4387319552 modified "2023-10-05" @default.
- W4387319552 title "Using Model Optimization as Countermeasure against Model Recovery Attacks" @default.
- W4387319552 cites W2154909745 @default.
- W4387319552 cites W2175377689 @default.
- W4387319552 cites W2194775991 @default.
- W4387319552 cites W2300242332 @default.
- W4387319552 cites W2511730936 @default.
- W4387319552 cites W2535690855 @default.
- W4387319552 cites W2928560789 @default.
- W4387319552 cites W2949650786 @default.
- W4387319552 cites W2982479999 @default.
- W4387319552 cites W3007566156 @default.
- W4387319552 cites W3035467254 @default.
- W4387319552 cites W3092516112 @default.
- W4387319552 cites W3134770507 @default.
- W4387319552 cites W3188169042 @default.
- W4387319552 cites W4300435436 @default.
- W4387319552 doi "https://doi.org/10.1007/978-3-031-41181-6_11" @default.
- W4387319552 hasPublicationYear "2023" @default.
- W4387319552 type Work @default.
- W4387319552 citedByCount "0" @default.
- W4387319552 crossrefType "book-chapter" @default.
- W4387319552 hasAuthorship W4387319552A5060395435 @default.
- W4387319552 hasAuthorship W4387319552A5081720773 @default.
- W4387319552 hasConcept C108010975 @default.
- W4387319552 hasConcept C108583219 @default.
- W4387319552 hasConcept C111919701 @default.
- W4387319552 hasConcept C115903868 @default.
- W4387319552 hasConcept C119857082 @default.
- W4387319552 hasConcept C138236772 @default.
- W4387319552 hasConcept C154945302 @default.
- W4387319552 hasConcept C162307627 @default.
- W4387319552 hasConcept C26713055 @default.
- W4387319552 hasConcept C2776214188 @default.
- W4387319552 hasConcept C38652104 @default.
- W4387319552 hasConcept C41008148 @default.
- W4387319552 hasConcept C50644808 @default.
- W4387319552 hasConcept C6557445 @default.
- W4387319552 hasConcept C65856478 @default.
- W4387319552 hasConcept C79974875 @default.
- W4387319552 hasConcept C86803240 @default.
- W4387319552 hasConcept C98045186 @default.
- W4387319552 hasConceptScore W4387319552C108010975 @default.
- W4387319552 hasConceptScore W4387319552C108583219 @default.
- W4387319552 hasConceptScore W4387319552C111919701 @default.
- W4387319552 hasConceptScore W4387319552C115903868 @default.
- W4387319552 hasConceptScore W4387319552C119857082 @default.
- W4387319552 hasConceptScore W4387319552C138236772 @default.
- W4387319552 hasConceptScore W4387319552C154945302 @default.
- W4387319552 hasConceptScore W4387319552C162307627 @default.
- W4387319552 hasConceptScore W4387319552C26713055 @default.
- W4387319552 hasConceptScore W4387319552C2776214188 @default.
- W4387319552 hasConceptScore W4387319552C38652104 @default.
- W4387319552 hasConceptScore W4387319552C41008148 @default.
- W4387319552 hasConceptScore W4387319552C50644808 @default.
- W4387319552 hasConceptScore W4387319552C6557445 @default.
- W4387319552 hasConceptScore W4387319552C65856478 @default.
- W4387319552 hasConceptScore W4387319552C79974875 @default.
- W4387319552 hasConceptScore W4387319552C86803240 @default.
- W4387319552 hasConceptScore W4387319552C98045186 @default.
- W4387319552 hasLocation W43873195521 @default.
- W4387319552 hasOpenAccess W4387319552 @default.
- W4387319552 hasPrimaryLocation W43873195521 @default.
- W4387319552 hasRelatedWork W2795261237 @default.
- W4387319552 hasRelatedWork W3014300295 @default.
- W4387319552 hasRelatedWork W3164822677 @default.
- W4387319552 hasRelatedWork W4223943233 @default.
- W4387319552 hasRelatedWork W4225161397 @default.
- W4387319552 hasRelatedWork W4312200629 @default.
- W4387319552 hasRelatedWork W4360585206 @default.
- W4387319552 hasRelatedWork W4364306694 @default.
- W4387319552 hasRelatedWork W4380075502 @default.
- W4387319552 hasRelatedWork W4380086463 @default.
- W4387319552 isParatext "false" @default.
- W4387319552 isRetracted "false" @default.
- W4387319552 workType "book-chapter" @default.