Matches in SemOpenAlex for { <https://semopenalex.org/work/W2910353502> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W2910353502 endingPage "191" @default.
- W2910353502 startingPage "174" @default.
- W2910353502 abstract "Neural Networks (NN) are today increasingly used in Machine Learning where they have become deeper and deeper to accurately model or classify high-level abstractions of data. Their development however also gives rise to important data privacy risks. This observation motives Microsoft researchers to propose a framework, called Cryptonets. The core idea is to combine simplifications of the NN with Fully Homomorphic Encryptions (FHE) techniques to get both confidentiality of the manipulated data and efficiency of the processing. While efficiency and accuracy are demonstrated when the number of non-linear layers is small (e.g. 2), Cryptonets unfortunately becomes ineffective for deeper NNs which let the privacy preserving problem open in these contexts. This work successfully addresses this problem by combining several new ideas including the use of the batch normalization principle and the splitting of the learning phase in several iterations. We experimentally validate the soundness of our approach with a neural network with 6 non-linear layers. When applied to the MNIST database, it competes with the accuracy of the best non-secure versions, thus significantly improving Cryptonets. Additionally, we applied our approach to secure a neural network used for face recognition. This problem is usually considered much harder than the MNIST hand-written digits recognition and can definitely not be addressed with a simple network like Cryptonets. By combining our new ideas with an iterative (learning) approach we experimentally show that we can build an FHE-friendly network achieving good accuracy for face recognition." @default.
- W2910353502 created "2019-01-25" @default.
- W2910353502 creator A5038985373 @default.
- W2910353502 creator A5060299604 @default.
- W2910353502 creator A5067956931 @default.
- W2910353502 creator A5078612546 @default.
- W2910353502 creator A5085840241 @default.
- W2910353502 date "2019-01-01" @default.
- W2910353502 modified "2023-10-17" @default.
- W2910353502 title "Recognition Over Encrypted Faces" @default.
- W2910353502 cites W133884053 @default.
- W2910353502 cites W1480376833 @default.
- W2910353502 cites W1544327602 @default.
- W2910353502 cites W1928278792 @default.
- W2910353502 cites W1971991172 @default.
- W2910353502 cites W2017464959 @default.
- W2910353502 cites W2024688311 @default.
- W2910353502 cites W2031738616 @default.
- W2910353502 cites W2077905990 @default.
- W2910353502 cites W2083847133 @default.
- W2910353502 cites W2137351756 @default.
- W2910353502 cites W2143612262 @default.
- W2910353502 cites W2155893237 @default.
- W2910353502 cites W2317339301 @default.
- W2910353502 cites W2474608001 @default.
- W2910353502 cites W2519249189 @default.
- W2910353502 cites W2701059868 @default.
- W2910353502 cites W2765200655 @default.
- W2910353502 cites W2963106566 @default.
- W2910353502 cites W3099206234 @default.
- W2910353502 cites W3141585064 @default.
- W2910353502 cites W4250776762 @default.
- W2910353502 doi "https://doi.org/10.1007/978-3-030-03101-5_16" @default.
- W2910353502 hasPublicationYear "2019" @default.
- W2910353502 type Work @default.
- W2910353502 sameAs 2910353502 @default.
- W2910353502 citedByCount "7" @default.
- W2910353502 countsByYear W29103535022020 @default.
- W2910353502 countsByYear W29103535022021 @default.
- W2910353502 countsByYear W29103535022023 @default.
- W2910353502 crossrefType "book-chapter" @default.
- W2910353502 hasAuthorship W2910353502A5038985373 @default.
- W2910353502 hasAuthorship W2910353502A5060299604 @default.
- W2910353502 hasAuthorship W2910353502A5067956931 @default.
- W2910353502 hasAuthorship W2910353502A5078612546 @default.
- W2910353502 hasAuthorship W2910353502A5085840241 @default.
- W2910353502 hasConcept C111472728 @default.
- W2910353502 hasConcept C119857082 @default.
- W2910353502 hasConcept C124101348 @default.
- W2910353502 hasConcept C136886441 @default.
- W2910353502 hasConcept C138885662 @default.
- W2910353502 hasConcept C144024400 @default.
- W2910353502 hasConcept C148730421 @default.
- W2910353502 hasConcept C153180895 @default.
- W2910353502 hasConcept C154945302 @default.
- W2910353502 hasConcept C158338273 @default.
- W2910353502 hasConcept C190502265 @default.
- W2910353502 hasConcept C19165224 @default.
- W2910353502 hasConcept C2780586882 @default.
- W2910353502 hasConcept C31510193 @default.
- W2910353502 hasConcept C38652104 @default.
- W2910353502 hasConcept C41008148 @default.
- W2910353502 hasConcept C50644808 @default.
- W2910353502 hasConceptScore W2910353502C111472728 @default.
- W2910353502 hasConceptScore W2910353502C119857082 @default.
- W2910353502 hasConceptScore W2910353502C124101348 @default.
- W2910353502 hasConceptScore W2910353502C136886441 @default.
- W2910353502 hasConceptScore W2910353502C138885662 @default.
- W2910353502 hasConceptScore W2910353502C144024400 @default.
- W2910353502 hasConceptScore W2910353502C148730421 @default.
- W2910353502 hasConceptScore W2910353502C153180895 @default.
- W2910353502 hasConceptScore W2910353502C154945302 @default.
- W2910353502 hasConceptScore W2910353502C158338273 @default.
- W2910353502 hasConceptScore W2910353502C190502265 @default.
- W2910353502 hasConceptScore W2910353502C19165224 @default.
- W2910353502 hasConceptScore W2910353502C2780586882 @default.
- W2910353502 hasConceptScore W2910353502C31510193 @default.
- W2910353502 hasConceptScore W2910353502C38652104 @default.
- W2910353502 hasConceptScore W2910353502C41008148 @default.
- W2910353502 hasConceptScore W2910353502C50644808 @default.
- W2910353502 hasLocation W29103535021 @default.
- W2910353502 hasOpenAccess W2910353502 @default.
- W2910353502 hasPrimaryLocation W29103535021 @default.
- W2910353502 hasRelatedWork W1863050261 @default.
- W2910353502 hasRelatedWork W1877610998 @default.
- W2910353502 hasRelatedWork W1991269640 @default.
- W2910353502 hasRelatedWork W2018445155 @default.
- W2910353502 hasRelatedWork W2071846378 @default.
- W2910353502 hasRelatedWork W2483947148 @default.
- W2910353502 hasRelatedWork W2910353502 @default.
- W2910353502 hasRelatedWork W2946038180 @default.
- W2910353502 hasRelatedWork W3149801936 @default.
- W2910353502 hasRelatedWork W2182569223 @default.
- W2910353502 isParatext "false" @default.
- W2910353502 isRetracted "false" @default.
- W2910353502 magId "2910353502" @default.
- W2910353502 workType "book-chapter" @default.