FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3118332057> ?p ?o ?g. }

• W3118332057 endingPage "560" @default.
• W3118332057 startingPage "535" @default.
• W3118332057 abstract "Although Convolutional Neural Networks (CNNs) are considered as being approximately invariant to nuisance perturbations such as image transformation, shift, scaling, and other small deformations, some existing studies show that intense noises can cause noticeable variation to CNNs’ outputs. This paper focuses on exploring a method of measuring sensitivity by observing corresponding output variation to input perturbation on CNNs. The sensitivity is statistically defined in a bottom-up way from neuron to layer, and finally to the entire CNN network. An iterative algorithm is proposed for approximating the defined sensitivity. On the basic architecture of CNNs, the theoretically computed sensitivity is verified on the MNIST database with four types of commonly used noise distributions: Gaussian, Uniform, Salt and Pepper, and Rayleigh. Experimental results show the theoretical sensitivity is on the one hand in agreement with the actual output variation what on the maps, layers or entire networks are, and on the other hand an applicable quantitative measure for robust network selection." @default.
• W3118332057 created "2021-01-18" @default.
• W3118332057 creator A5008783473 @default.
• W3118332057 creator A5063853753 @default.
• W3118332057 creator A5069016585 @default.
• W3118332057 creator A5072246520 @default.
• W3118332057 creator A5088665887 @default.
• W3118332057 date "2021-01-03" @default.
• W3118332057 modified "2023-10-04" @default.
• W3118332057 title "Computation of CNN’s Sensitivity to Input Perturbation" @default.
• W3118332057 cites W1495464107 @default.
• W3118332057 cites W1496650988 @default.
• W3118332057 cites W1948751323 @default.
• W3118332057 cites W1976841235 @default.
• W3118332057 cites W2008870058 @default.
• W3118332057 cites W2014181466 @default.
• W3118332057 cites W2016053056 @default.
• W3118332057 cites W2101241495 @default.
• W3118332057 cites W2102605133 @default.
• W3118332057 cites W2106076725 @default.
• W3118332057 cites W2107808739 @default.
• W3118332057 cites W2110480233 @default.
• W3118332057 cites W2114027591 @default.
• W3118332057 cites W2139499553 @default.
• W3118332057 cites W2159778339 @default.
• W3118332057 cites W2160690336 @default.
• W3118332057 cites W2167677193 @default.
• W3118332057 cites W2167901303 @default.
• W3118332057 cites W2169892349 @default.
• W3118332057 cites W2535873859 @default.
• W3118332057 cites W2575696673 @default.
• W3118332057 cites W2622826443 @default.
• W3118332057 cites W2767471303 @default.
• W3118332057 cites W2962700793 @default.
• W3118332057 cites W2962804139 @default.
• W3118332057 cites W2963324806 @default.
• W3118332057 cites W2963669006 @default.
• W3118332057 cites W2963739340 @default.
• W3118332057 cites W2971902683 @default.
• W3118332057 cites W2994265419 @default.
• W3118332057 cites W3007993482 @default.
• W3118332057 cites W3013105413 @default.
• W3118332057 cites W3014858325 @default.
• W3118332057 cites W3029252806 @default.
• W3118332057 cites W3032667095 @default.
• W3118332057 cites W3087312110 @default.
• W3118332057 doi "https://doi.org/10.1007/s11063-020-10420-7" @default.
• W3118332057 hasPublicationYear "2021" @default.
• W3118332057 type Work @default.
• W3118332057 sameAs 3118332057 @default.
• W3118332057 citedByCount "4" @default.
• W3118332057 countsByYear W31183320572021 @default.
• W3118332057 countsByYear W31183320572022 @default.
• W3118332057 countsByYear W31183320572023 @default.
• W3118332057 crossrefType "journal-article" @default.
• W3118332057 hasAuthorship W3118332057A5008783473 @default.
• W3118332057 hasAuthorship W3118332057A5063853753 @default.
• W3118332057 hasAuthorship W3118332057A5069016585 @default.
• W3118332057 hasAuthorship W3118332057A5072246520 @default.
• W3118332057 hasAuthorship W3118332057A5088665887 @default.
• W3118332057 hasBestOaLocation W31183320572 @default.
• W3118332057 hasConcept C11413529 @default.
• W3118332057 hasConcept C121332964 @default.
• W3118332057 hasConcept C127413603 @default.
• W3118332057 hasConcept C153180895 @default.
• W3118332057 hasConcept C154945302 @default.
• W3118332057 hasConcept C163716315 @default.
• W3118332057 hasConcept C177918212 @default.
• W3118332057 hasConcept C190470478 @default.
• W3118332057 hasConcept C190502265 @default.
• W3118332057 hasConcept C21200559 @default.
• W3118332057 hasConcept C24326235 @default.
• W3118332057 hasConcept C2524010 @default.
• W3118332057 hasConcept C33923547 @default.
• W3118332057 hasConcept C37914503 @default.
• W3118332057 hasConcept C41008148 @default.
• W3118332057 hasConcept C45374587 @default.
• W3118332057 hasConcept C50644808 @default.
• W3118332057 hasConcept C62520636 @default.
• W3118332057 hasConcept C81363708 @default.
• W3118332057 hasConcept C99844830 @default.
• W3118332057 hasConceptScore W3118332057C11413529 @default.
• W3118332057 hasConceptScore W3118332057C121332964 @default.
• W3118332057 hasConceptScore W3118332057C127413603 @default.
• W3118332057 hasConceptScore W3118332057C153180895 @default.
• W3118332057 hasConceptScore W3118332057C154945302 @default.
• W3118332057 hasConceptScore W3118332057C163716315 @default.
• W3118332057 hasConceptScore W3118332057C177918212 @default.
• W3118332057 hasConceptScore W3118332057C190470478 @default.
• W3118332057 hasConceptScore W3118332057C190502265 @default.
• W3118332057 hasConceptScore W3118332057C21200559 @default.
• W3118332057 hasConceptScore W3118332057C24326235 @default.
• W3118332057 hasConceptScore W3118332057C2524010 @default.
• W3118332057 hasConceptScore W3118332057C33923547 @default.
• W3118332057 hasConceptScore W3118332057C37914503 @default.
• W3118332057 hasConceptScore W3118332057C41008148 @default.
• W3118332057 hasConceptScore W3118332057C45374587 @default.
• W3118332057 hasConceptScore W3118332057C50644808 @default.

• next