Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308288352> ?p ?o ?g. }
Showing items 1 to 62 of
62
with 100 items per page.
- W4308288352 abstract "How can we bring both privacy and energy-efficiency to a neural system? In this paper, we propose PrivateSNN, which aims to build low-power Spiking Neural Networks (SNNs) from a pre-trained ANN model without leaking sensitive information contained in a dataset. Here, we tackle two types of leakage problems: 1) Data leakage is caused when the networks access real training data during an ANN-SNN conversion process. 2) Class leakage is caused when class-related features can be reconstructed from network parameters. In order to address the data leakage issue, we generate synthetic images from the pre-trained ANNs and convert ANNs to SNNs using the generated images. However, converted SNNs remain vulnerable to class leakage since the weight parameters have the same (or scaled) value with respect to ANN parameters. Therefore, we encrypt SNN weights by training SNNs with a temporal spike-based learning rule. Updating weight parameters with temporal data makes SNNs difficult to be interpreted in the spatial domain. We observe that the encrypted PrivateSNN eliminates data and class leakage issues with a slight performance drop (less than ~2) and significant energy-efficiency gain (about 55x) compared to the standard ANN. We conduct extensive experiments on various datasets including CIFAR10, CIFAR100, and TinyImageNet, highlighting the importance of privacy-preserving SNN training." @default.
- W4308288352 created "2022-11-10" @default.
- W4308288352 creator A5050310538 @default.
- W4308288352 creator A5079036687 @default.
- W4308288352 creator A5086829943 @default.
- W4308288352 date "2021-04-07" @default.
- W4308288352 modified "2023-09-28" @default.
- W4308288352 title "PrivateSNN: Privacy-Preserving Spiking Neural Networks" @default.
- W4308288352 doi "https://doi.org/10.48550/arxiv.2104.03414" @default.
- W4308288352 hasPublicationYear "2021" @default.
- W4308288352 type Work @default.
- W4308288352 citedByCount "0" @default.
- W4308288352 crossrefType "posted-content" @default.
- W4308288352 hasAuthorship W4308288352A5050310538 @default.
- W4308288352 hasAuthorship W4308288352A5079036687 @default.
- W4308288352 hasAuthorship W4308288352A5086829943 @default.
- W4308288352 hasBestOaLocation W43082883521 @default.
- W4308288352 hasConcept C11731999 @default.
- W4308288352 hasConcept C119857082 @default.
- W4308288352 hasConcept C124101348 @default.
- W4308288352 hasConcept C139719470 @default.
- W4308288352 hasConcept C148730421 @default.
- W4308288352 hasConcept C153180895 @default.
- W4308288352 hasConcept C154945302 @default.
- W4308288352 hasConcept C162324750 @default.
- W4308288352 hasConcept C2777042071 @default.
- W4308288352 hasConcept C2777212361 @default.
- W4308288352 hasConcept C2779201187 @default.
- W4308288352 hasConcept C31258907 @default.
- W4308288352 hasConcept C41008148 @default.
- W4308288352 hasConcept C50644808 @default.
- W4308288352 hasConceptScore W4308288352C11731999 @default.
- W4308288352 hasConceptScore W4308288352C119857082 @default.
- W4308288352 hasConceptScore W4308288352C124101348 @default.
- W4308288352 hasConceptScore W4308288352C139719470 @default.
- W4308288352 hasConceptScore W4308288352C148730421 @default.
- W4308288352 hasConceptScore W4308288352C153180895 @default.
- W4308288352 hasConceptScore W4308288352C154945302 @default.
- W4308288352 hasConceptScore W4308288352C162324750 @default.
- W4308288352 hasConceptScore W4308288352C2777042071 @default.
- W4308288352 hasConceptScore W4308288352C2777212361 @default.
- W4308288352 hasConceptScore W4308288352C2779201187 @default.
- W4308288352 hasConceptScore W4308288352C31258907 @default.
- W4308288352 hasConceptScore W4308288352C41008148 @default.
- W4308288352 hasConceptScore W4308288352C50644808 @default.
- W4308288352 hasLocation W43082883521 @default.
- W4308288352 hasLocation W43082883522 @default.
- W4308288352 hasOpenAccess W4308288352 @default.
- W4308288352 hasPrimaryLocation W43082883521 @default.
- W4308288352 hasRelatedWork W2004039463 @default.
- W4308288352 hasRelatedWork W2161954659 @default.
- W4308288352 hasRelatedWork W2290623933 @default.
- W4308288352 hasRelatedWork W2362737126 @default.
- W4308288352 hasRelatedWork W2885843018 @default.
- W4308288352 hasRelatedWork W2961085424 @default.
- W4308288352 hasRelatedWork W4306674287 @default.
- W4308288352 hasRelatedWork W4308432025 @default.
- W4308288352 hasRelatedWork W1629725936 @default.
- W4308288352 hasRelatedWork W4224009465 @default.
- W4308288352 isParatext "false" @default.
- W4308288352 isRetracted "false" @default.
- W4308288352 workType "article" @default.