Matches in SemOpenAlex for { <https://semopenalex.org/work/W2797333460> ?p ?o ?g. }
- W2797333460 abstract "This paper introduces Non-Autonomous Input-Output Stable Network(NAIS-Net), a very deep architecture where each stacked processing block is derived from a time-invariant non-autonomous dynamical system. Non-autonomy is implemented by skip connections from the block input to each of the unrolled processing stages and allows stability to be enforced so that blocks can be unrolled adaptively to a pattern-dependent processing depth. NAIS-Net induces non-trivial, Lipschitz input-output maps, even for an infinite unroll length. We prove that the network is globally asymptotically stable so that for every initial condition there is exactly one input-dependent equilibrium assuming $tanh$ units, and incrementally stable for ReL units. An efficient implementation that enforces the stability under derived conditions for both fully-connected and convolutional layers is also presented. Experimental results show how NAIS-Net exhibits stability in practice, yielding a significant reduction in generalization gap compared to ResNets." @default.
- W2797333460 created "2018-04-24" @default.
- W2797333460 creator A5024791802 @default.
- W2797333460 creator A5040835177 @default.
- W2797333460 creator A5046097198 @default.
- W2797333460 creator A5047652078 @default.
- W2797333460 creator A5068886220 @default.
- W2797333460 date "2018-04-19" @default.
- W2797333460 modified "2023-09-28" @default.
- W2797333460 title "NAIS-Net: Stable Deep Networks from Non-Autonomous Differential Equations" @default.
- W2797333460 cites W1484739396 @default.
- W2797333460 cites W154005121 @default.
- W2797333460 cites W1597944220 @default.
- W2797333460 cites W1631735240 @default.
- W2797333460 cites W2029030701 @default.
- W2797333460 cites W2064675550 @default.
- W2797333460 cites W2097311594 @default.
- W2797333460 cites W2108598243 @default.
- W2797333460 cites W2118103795 @default.
- W2797333460 cites W2130942839 @default.
- W2797333460 cites W2135263068 @default.
- W2797333460 cites W2144001972 @default.
- W2797333460 cites W2157331557 @default.
- W2797333460 cites W2178103884 @default.
- W2797333460 cites W2194775991 @default.
- W2797333460 cites W2325237720 @default.
- W2797333460 cites W2337199865 @default.
- W2797333460 cites W2408279554 @default.
- W2797333460 cites W2558460151 @default.
- W2797333460 cites W2565532197 @default.
- W2797333460 cites W2587753047 @default.
- W2797333460 cites W2600297185 @default.
- W2797333460 cites W2610190180 @default.
- W2797333460 cites W2610857016 @default.
- W2797333460 cites W2618017917 @default.
- W2797333460 cites W2695777493 @default.
- W2797333460 cites W2739427748 @default.
- W2797333460 cites W2752813595 @default.
- W2797333460 cites W2765720631 @default.
- W2797333460 cites W2766207105 @default.
- W2797333460 cites W2772087644 @default.
- W2797333460 cites W2774627531 @default.
- W2797333460 cites W2786406400 @default.
- W2797333460 cites W2949117887 @default.
- W2797333460 cites W2949242443 @default.
- W2797333460 cites W2949608135 @default.
- W2797333460 cites W2951985269 @default.
- W2797333460 cites W2962737770 @default.
- W2797333460 cites W2963302407 @default.
- W2797333460 cites W2963446712 @default.
- W2797333460 cites W2963684275 @default.
- W2797333460 cites W2963836885 @default.
- W2797333460 cites W2963983719 @default.
- W2797333460 cites W2964153729 @default.
- W2797333460 cites W3118608800 @default.
- W2797333460 hasPublicationYear "2018" @default.
- W2797333460 type Work @default.
- W2797333460 sameAs 2797333460 @default.
- W2797333460 citedByCount "16" @default.
- W2797333460 countsByYear W27973334602018 @default.
- W2797333460 countsByYear W27973334602019 @default.
- W2797333460 countsByYear W27973334602020 @default.
- W2797333460 countsByYear W27973334602021 @default.
- W2797333460 crossrefType "posted-content" @default.
- W2797333460 hasAuthorship W2797333460A5024791802 @default.
- W2797333460 hasAuthorship W2797333460A5040835177 @default.
- W2797333460 hasAuthorship W2797333460A5046097198 @default.
- W2797333460 hasAuthorship W2797333460A5047652078 @default.
- W2797333460 hasAuthorship W2797333460A5068886220 @default.
- W2797333460 hasConcept C111335779 @default.
- W2797333460 hasConcept C112972136 @default.
- W2797333460 hasConcept C114614502 @default.
- W2797333460 hasConcept C119857082 @default.
- W2797333460 hasConcept C134306372 @default.
- W2797333460 hasConcept C14166107 @default.
- W2797333460 hasConcept C154945302 @default.
- W2797333460 hasConcept C177148314 @default.
- W2797333460 hasConcept C184720557 @default.
- W2797333460 hasConcept C190470478 @default.
- W2797333460 hasConcept C202444582 @default.
- W2797333460 hasConcept C22324862 @default.
- W2797333460 hasConcept C2524010 @default.
- W2797333460 hasConcept C2775924081 @default.
- W2797333460 hasConcept C2777210771 @default.
- W2797333460 hasConcept C33923547 @default.
- W2797333460 hasConcept C37914503 @default.
- W2797333460 hasConcept C41008148 @default.
- W2797333460 hasConcept C47446073 @default.
- W2797333460 hasConceptScore W2797333460C111335779 @default.
- W2797333460 hasConceptScore W2797333460C112972136 @default.
- W2797333460 hasConceptScore W2797333460C114614502 @default.
- W2797333460 hasConceptScore W2797333460C119857082 @default.
- W2797333460 hasConceptScore W2797333460C134306372 @default.
- W2797333460 hasConceptScore W2797333460C14166107 @default.
- W2797333460 hasConceptScore W2797333460C154945302 @default.
- W2797333460 hasConceptScore W2797333460C177148314 @default.
- W2797333460 hasConceptScore W2797333460C184720557 @default.
- W2797333460 hasConceptScore W2797333460C190470478 @default.
- W2797333460 hasConceptScore W2797333460C202444582 @default.
- W2797333460 hasConceptScore W2797333460C22324862 @default.