Matches in SemOpenAlex for { <https://semopenalex.org/work/W3088690503> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3088690503 abstract "In this work, we quantify how neural networks designed from biology using no statistical training have a remarkable performance in information theoretic terms. Specifically, we address the question of the amount of information that can be extracted about the images from the different layers of psychophysically tuned deep networks. We show that analytical approaches are not possible, and we propose the use of two empirical estimators of capacity: the classical Kozachenko-Lonenko estimator and a recent estimator based on Gaussianization. Results show that networks purely based on visual psychophysics are extremely efficient in two aspects: (1) the internal representation of these networks duplicates the amount of information that can be extracted about the images with regard to the amount of information that could be obtained from the input representation assuming sensors of the same quality, and (2) the capacity of internal representation follows the PDF of natural scenes over the chromatic and achromatic dimensions of the stimulus space. This remarkable adaptation to the natural environment is an example of how imitation of biological vision may inspire architectures and save training effort in artificial vision." @default.
- W3088690503 created "2020-10-01" @default.
- W3088690503 creator A5065733630 @default.
- W3088690503 date "2020-09-22" @default.
- W3088690503 modified "2023-09-23" @default.
- W3088690503 title "Channel Capacity in Psychovisual Deep-Nets: Gaussianization Versus Kozachenko-Leonenko" @default.
- W3088690503 cites W1520494951 @default.
- W3088690503 cites W1972348417 @default.
- W3088690503 cites W1987125655 @default.
- W3088690503 cites W1988013646 @default.
- W3088690503 cites W1992546932 @default.
- W3088690503 cites W2000674655 @default.
- W3088690503 cites W2046119925 @default.
- W3088690503 cites W2064703793 @default.
- W3088690503 cites W2121914520 @default.
- W3088690503 cites W2123566137 @default.
- W3088690503 cites W2130055919 @default.
- W3088690503 cites W2143612128 @default.
- W3088690503 cites W2166078937 @default.
- W3088690503 cites W2166278041 @default.
- W3088690503 cites W2502575609 @default.
- W3088690503 cites W2581944749 @default.
- W3088690503 cites W3098585936 @default.
- W3088690503 cites W3103805372 @default.
- W3088690503 cites W3012261518 @default.
- W3088690503 doi "https://doi.org/10.1007/978-981-15-6067-5_10" @default.
- W3088690503 hasPublicationYear "2020" @default.
- W3088690503 type Work @default.
- W3088690503 sameAs 3088690503 @default.
- W3088690503 citedByCount "0" @default.
- W3088690503 crossrefType "book-chapter" @default.
- W3088690503 hasAuthorship W3088690503A5065733630 @default.
- W3088690503 hasConcept C105795698 @default.
- W3088690503 hasConcept C114614502 @default.
- W3088690503 hasConcept C119857082 @default.
- W3088690503 hasConcept C121332964 @default.
- W3088690503 hasConcept C121535716 @default.
- W3088690503 hasConcept C1276947 @default.
- W3088690503 hasConcept C15123163 @default.
- W3088690503 hasConcept C153180895 @default.
- W3088690503 hasConcept C154945302 @default.
- W3088690503 hasConcept C169760540 @default.
- W3088690503 hasConcept C17744445 @default.
- W3088690503 hasConcept C185429906 @default.
- W3088690503 hasConcept C196956537 @default.
- W3088690503 hasConcept C199539241 @default.
- W3088690503 hasConcept C26760741 @default.
- W3088690503 hasConcept C2776359362 @default.
- W3088690503 hasConcept C31972630 @default.
- W3088690503 hasConcept C33923547 @default.
- W3088690503 hasConcept C41008148 @default.
- W3088690503 hasConcept C50644808 @default.
- W3088690503 hasConcept C52622258 @default.
- W3088690503 hasConcept C80444323 @default.
- W3088690503 hasConcept C86803240 @default.
- W3088690503 hasConcept C94625758 @default.
- W3088690503 hasConceptScore W3088690503C105795698 @default.
- W3088690503 hasConceptScore W3088690503C114614502 @default.
- W3088690503 hasConceptScore W3088690503C119857082 @default.
- W3088690503 hasConceptScore W3088690503C121332964 @default.
- W3088690503 hasConceptScore W3088690503C121535716 @default.
- W3088690503 hasConceptScore W3088690503C1276947 @default.
- W3088690503 hasConceptScore W3088690503C15123163 @default.
- W3088690503 hasConceptScore W3088690503C153180895 @default.
- W3088690503 hasConceptScore W3088690503C154945302 @default.
- W3088690503 hasConceptScore W3088690503C169760540 @default.
- W3088690503 hasConceptScore W3088690503C17744445 @default.
- W3088690503 hasConceptScore W3088690503C185429906 @default.
- W3088690503 hasConceptScore W3088690503C196956537 @default.
- W3088690503 hasConceptScore W3088690503C199539241 @default.
- W3088690503 hasConceptScore W3088690503C26760741 @default.
- W3088690503 hasConceptScore W3088690503C2776359362 @default.
- W3088690503 hasConceptScore W3088690503C31972630 @default.
- W3088690503 hasConceptScore W3088690503C33923547 @default.
- W3088690503 hasConceptScore W3088690503C41008148 @default.
- W3088690503 hasConceptScore W3088690503C50644808 @default.
- W3088690503 hasConceptScore W3088690503C52622258 @default.
- W3088690503 hasConceptScore W3088690503C80444323 @default.
- W3088690503 hasConceptScore W3088690503C86803240 @default.
- W3088690503 hasConceptScore W3088690503C94625758 @default.
- W3088690503 hasLocation W30886905031 @default.
- W3088690503 hasOpenAccess W3088690503 @default.
- W3088690503 hasPrimaryLocation W30886905031 @default.
- W3088690503 hasRelatedWork W11538522 @default.
- W3088690503 hasRelatedWork W14195083 @default.
- W3088690503 hasRelatedWork W1562032 @default.
- W3088690503 hasRelatedWork W2966661 @default.
- W3088690503 hasRelatedWork W3279675 @default.
- W3088690503 hasRelatedWork W7283634 @default.
- W3088690503 hasRelatedWork W7497093 @default.
- W3088690503 hasRelatedWork W7657724 @default.
- W3088690503 hasRelatedWork W9321062 @default.
- W3088690503 hasRelatedWork W9770290 @default.
- W3088690503 isParatext "false" @default.
- W3088690503 isRetracted "false" @default.
- W3088690503 magId "3088690503" @default.
- W3088690503 workType "book-chapter" @default.