Matches in SemOpenAlex for { <https://semopenalex.org/work/W2972182854> ?p ?o ?g. }
- W2972182854 endingPage "433" @default.
- W2972182854 startingPage "423" @default.
- W2972182854 abstract "We provide a general mathematical framework for group and set equivariance in machine learning. We define group equivariant non-expansive operators (GENEOs) as maps between function spaces associated with groups of transformations. We study the topological and metric properties of the space of GENEOs to evaluate their approximating power and set the basis for general strategies to initialize and compose operators. We define suitable pseudo-metrics for the function spaces, the equivariance groups and the set of non-expansive operators. We prove that, under suitable assumptions, the space of GENEOs is compact and convex. These results provide fundamental guarantees in a machine learning perspective. By considering isometry-equivariant non-expansive operators, we describe a simple strategy to select and sample operators. Thereafter, we show how selected and sampled operators can be used both to perform classical metric learning and to inject knowledge in artificial neural networks. Controlling the flow and representation of information in deep neural networks is fundamental to making networks intelligible. Bergomi et al introduce a mathematical framework in which the space of possible operators representing the data is constrained by using symmetries. This constrained space is still suitable for machine learning: operators can be efficiently computed, approximated and parameterized for optimization." @default.
- W2972182854 created "2019-09-12" @default.
- W2972182854 creator A5010581059 @default.
- W2972182854 creator A5045329472 @default.
- W2972182854 creator A5057632632 @default.
- W2972182854 creator A5058035420 @default.
- W2972182854 date "2019-09-02" @default.
- W2972182854 modified "2023-10-14" @default.
- W2972182854 title "Towards a topological–geometrical theory of group equivariant non-expansive operators for data analysis and machine learning" @default.
- W2972182854 cites W1635792279 @default.
- W2972182854 cites W1756438045 @default.
- W2972182854 cites W1832693441 @default.
- W2972182854 cites W1977758817 @default.
- W2972182854 cites W1979273469 @default.
- W2972182854 cites W1989757660 @default.
- W2972182854 cites W1991566301 @default.
- W2972182854 cites W1994906459 @default.
- W2972182854 cites W2007339694 @default.
- W2972182854 cites W2014209644 @default.
- W2972182854 cites W2020556227 @default.
- W2972182854 cites W2024915064 @default.
- W2972182854 cites W2053798192 @default.
- W2972182854 cites W2105732809 @default.
- W2972182854 cites W2156247618 @default.
- W2972182854 cites W2163922914 @default.
- W2972182854 cites W2261689926 @default.
- W2972182854 cites W2399437674 @default.
- W2972182854 cites W2576915720 @default.
- W2972182854 cites W2581082771 @default.
- W2972182854 cites W2622826443 @default.
- W2972182854 cites W2766447205 @default.
- W2972182854 cites W2886960264 @default.
- W2972182854 cites W2964184826 @default.
- W2972182854 cites W2964253930 @default.
- W2972182854 cites W2966661 @default.
- W2972182854 cites W3123784055 @default.
- W2972182854 cites W4233690291 @default.
- W2972182854 cites W4251980118 @default.
- W2972182854 doi "https://doi.org/10.1038/s42256-019-0087-3" @default.
- W2972182854 hasPublicationYear "2019" @default.
- W2972182854 type Work @default.
- W2972182854 sameAs 2972182854 @default.
- W2972182854 citedByCount "26" @default.
- W2972182854 countsByYear W29721828542019 @default.
- W2972182854 countsByYear W29721828542020 @default.
- W2972182854 countsByYear W29721828542021 @default.
- W2972182854 countsByYear W29721828542022 @default.
- W2972182854 countsByYear W29721828542023 @default.
- W2972182854 crossrefType "journal-article" @default.
- W2972182854 hasAuthorship W2972182854A5010581059 @default.
- W2972182854 hasAuthorship W2972182854A5045329472 @default.
- W2972182854 hasAuthorship W2972182854A5057632632 @default.
- W2972182854 hasAuthorship W2972182854A5058035420 @default.
- W2972182854 hasBestOaLocation W29721828542 @default.
- W2972182854 hasConcept C11413529 @default.
- W2972182854 hasConcept C114614502 @default.
- W2972182854 hasConcept C14036430 @default.
- W2972182854 hasConcept C154945302 @default.
- W2972182854 hasConcept C162324750 @default.
- W2972182854 hasConcept C165464430 @default.
- W2972182854 hasConcept C171036898 @default.
- W2972182854 hasConcept C176217482 @default.
- W2972182854 hasConcept C177264268 @default.
- W2972182854 hasConcept C17744445 @default.
- W2972182854 hasConcept C184720557 @default.
- W2972182854 hasConcept C199360897 @default.
- W2972182854 hasConcept C199539241 @default.
- W2972182854 hasConcept C202444582 @default.
- W2972182854 hasConcept C21547014 @default.
- W2972182854 hasConcept C2776359362 @default.
- W2972182854 hasConcept C33923547 @default.
- W2972182854 hasConcept C41008148 @default.
- W2972182854 hasConcept C50644808 @default.
- W2972182854 hasConcept C78458016 @default.
- W2972182854 hasConcept C86803240 @default.
- W2972182854 hasConcept C94625758 @default.
- W2972182854 hasConceptScore W2972182854C11413529 @default.
- W2972182854 hasConceptScore W2972182854C114614502 @default.
- W2972182854 hasConceptScore W2972182854C14036430 @default.
- W2972182854 hasConceptScore W2972182854C154945302 @default.
- W2972182854 hasConceptScore W2972182854C162324750 @default.
- W2972182854 hasConceptScore W2972182854C165464430 @default.
- W2972182854 hasConceptScore W2972182854C171036898 @default.
- W2972182854 hasConceptScore W2972182854C176217482 @default.
- W2972182854 hasConceptScore W2972182854C177264268 @default.
- W2972182854 hasConceptScore W2972182854C17744445 @default.
- W2972182854 hasConceptScore W2972182854C184720557 @default.
- W2972182854 hasConceptScore W2972182854C199360897 @default.
- W2972182854 hasConceptScore W2972182854C199539241 @default.
- W2972182854 hasConceptScore W2972182854C202444582 @default.
- W2972182854 hasConceptScore W2972182854C21547014 @default.
- W2972182854 hasConceptScore W2972182854C2776359362 @default.
- W2972182854 hasConceptScore W2972182854C33923547 @default.
- W2972182854 hasConceptScore W2972182854C41008148 @default.
- W2972182854 hasConceptScore W2972182854C50644808 @default.
- W2972182854 hasConceptScore W2972182854C78458016 @default.
- W2972182854 hasConceptScore W2972182854C86803240 @default.
- W2972182854 hasConceptScore W2972182854C94625758 @default.