Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891497957> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2891497957 abstract "Deep neural networks have become the main work horse for many tasks involving learning from data in a variety of applications in Science and Engineering. Traditionally, the input to these networks lie in a vector space and the operations employed within the network are well defined on vector-spaces. In the recent past, due to technological advances in sensing, it has become possible to acquire manifold-valued data sets either directly or indirectly. Examples include but are not limited to data from omnidirectional cameras on automobiles, drones etc., synthetic aperture radar imaging, diffusion magnetic resonance imaging, elastography and conductance imaging in the Medical Imaging domain and others. Thus, there is need to generalize the deep neural networks to cope with input data that reside on curved manifolds where vector space operations are not naturally admissible. In this paper, we present a novel theoretical framework to generalize the widely popular convolutional neural networks (CNNs) to high dimensional manifold-valued data inputs. We call these networks, ManifoldNets. In ManifoldNets, convolution operation on data residing on Riemannian manifolds is achieved via a provably convergent recursive computation of the weighted Frechet Mean (wFM) of the given data, where the weights makeup the convolution mask, to be learned. Further, we prove that the proposed wFM layer achieves a contraction mapping and hence ManifoldNet does not need the non-linear ReLU unit used in standard CNNs. We present experiments, using the ManifoldNet framework, to achieve dimensionality reduction by computing the principal linear subspaces that naturally reside on a Grassmannian. The experimental results demonstrate the efficacy of ManifoldNets in the context of classification and reconstruction accuracy." @default.
- W2891497957 created "2018-09-27" @default.
- W2891497957 creator A5016378470 @default.
- W2891497957 creator A5025100345 @default.
- W2891497957 creator A5039442646 @default.
- W2891497957 creator A5046301126 @default.
- W2891497957 date "2018-09-11" @default.
- W2891497957 modified "2023-09-27" @default.
- W2891497957 title "ManifoldNet: A Deep Network Framework for Manifold-valued Data" @default.
- W2891497957 cites W2163922914 @default.
- W2891497957 cites W2551687757 @default.
- W2891497957 cites W2751473119 @default.
- W2891497957 cites W2785397276 @default.
- W2891497957 cites W2804126069 @default.
- W2891497957 cites W2952054889 @default.
- W2891497957 cites W2952560675 @default.
- W2891497957 cites W3129186208 @default.
- W2891497957 cites W637153065 @default.
- W2891497957 hasPublicationYear "2018" @default.
- W2891497957 type Work @default.
- W2891497957 sameAs 2891497957 @default.
- W2891497957 citedByCount "11" @default.
- W2891497957 countsByYear W28914979572019 @default.
- W2891497957 countsByYear W28914979572020 @default.
- W2891497957 countsByYear W28914979572021 @default.
- W2891497957 crossrefType "posted-content" @default.
- W2891497957 hasAuthorship W2891497957A5016378470 @default.
- W2891497957 hasAuthorship W2891497957A5025100345 @default.
- W2891497957 hasAuthorship W2891497957A5039442646 @default.
- W2891497957 hasAuthorship W2891497957A5046301126 @default.
- W2891497957 hasConcept C108583219 @default.
- W2891497957 hasConcept C127413603 @default.
- W2891497957 hasConcept C153180895 @default.
- W2891497957 hasConcept C154945302 @default.
- W2891497957 hasConcept C202444582 @default.
- W2891497957 hasConcept C2779593128 @default.
- W2891497957 hasConcept C33923547 @default.
- W2891497957 hasConcept C41008148 @default.
- W2891497957 hasConcept C50644808 @default.
- W2891497957 hasConcept C529865628 @default.
- W2891497957 hasConcept C70518039 @default.
- W2891497957 hasConcept C78519656 @default.
- W2891497957 hasConcept C81363708 @default.
- W2891497957 hasConceptScore W2891497957C108583219 @default.
- W2891497957 hasConceptScore W2891497957C127413603 @default.
- W2891497957 hasConceptScore W2891497957C153180895 @default.
- W2891497957 hasConceptScore W2891497957C154945302 @default.
- W2891497957 hasConceptScore W2891497957C202444582 @default.
- W2891497957 hasConceptScore W2891497957C2779593128 @default.
- W2891497957 hasConceptScore W2891497957C33923547 @default.
- W2891497957 hasConceptScore W2891497957C41008148 @default.
- W2891497957 hasConceptScore W2891497957C50644808 @default.
- W2891497957 hasConceptScore W2891497957C529865628 @default.
- W2891497957 hasConceptScore W2891497957C70518039 @default.
- W2891497957 hasConceptScore W2891497957C78519656 @default.
- W2891497957 hasConceptScore W2891497957C81363708 @default.
- W2891497957 hasLocation W28914979571 @default.
- W2891497957 hasOpenAccess W2891497957 @default.
- W2891497957 hasPrimaryLocation W28914979571 @default.
- W2891497957 hasRelatedWork W1037879298 @default.
- W2891497957 hasRelatedWork W1991605728 @default.
- W2891497957 hasRelatedWork W2064675550 @default.
- W2891497957 hasRelatedWork W2194775991 @default.
- W2891497957 hasRelatedWork W2217701558 @default.
- W2891497957 hasRelatedWork W2558748708 @default.
- W2891497957 hasRelatedWork W2604053529 @default.
- W2891497957 hasRelatedWork W2770865047 @default.
- W2891497957 hasRelatedWork W2783160710 @default.
- W2891497957 hasRelatedWork W2785397276 @default.
- W2891497957 hasRelatedWork W2803706486 @default.
- W2891497957 hasRelatedWork W2804126069 @default.
- W2891497957 hasRelatedWork W2962863013 @default.
- W2891497957 hasRelatedWork W2963689654 @default.
- W2891497957 hasRelatedWork W2964330148 @default.
- W2891497957 hasRelatedWork W3015069144 @default.
- W2891497957 hasRelatedWork W3036536899 @default.
- W2891497957 hasRelatedWork W3046455422 @default.
- W2891497957 hasRelatedWork W3196558741 @default.
- W2891497957 hasRelatedWork W3204755622 @default.
- W2891497957 isParatext "false" @default.
- W2891497957 isRetracted "false" @default.
- W2891497957 magId "2891497957" @default.
- W2891497957 workType "article" @default.