Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287554868> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4287554868 abstract "Neural networks have shown tremendous potential for reconstructing high-resolution images in inverse problems. The non-convex and opaque nature of neural networks, however, hinders their utility in sensitive applications such as medical imaging. To cope with this challenge, this paper advocates a convex duality framework that makes a two-layer fully-convolutional ReLU denoising network amenable to convex optimization. The convex dual network not only offers the optimum training with convex solvers, but also facilitates interpreting training and prediction. In particular, it implies training neural networks with weight decay regularization induces path sparsity while the prediction is piecewise linear filtering. A range of experiments with MNIST and fastMRI datasets confirm the efficacy of the dual network optimization problem." @default.
- W4287554868 created "2022-07-25" @default.
- W4287554868 creator A5001436196 @default.
- W4287554868 creator A5008348052 @default.
- W4287554868 creator A5030024620 @default.
- W4287554868 creator A5060755739 @default.
- W4287554868 creator A5072391499 @default.
- W4287554868 date "2020-12-09" @default.
- W4287554868 modified "2023-09-27" @default.
- W4287554868 title "Convex Regularization Behind Neural Reconstruction" @default.
- W4287554868 doi "https://doi.org/10.48550/arxiv.2012.05169" @default.
- W4287554868 hasPublicationYear "2020" @default.
- W4287554868 type Work @default.
- W4287554868 citedByCount "0" @default.
- W4287554868 crossrefType "posted-content" @default.
- W4287554868 hasAuthorship W4287554868A5001436196 @default.
- W4287554868 hasAuthorship W4287554868A5008348052 @default.
- W4287554868 hasAuthorship W4287554868A5030024620 @default.
- W4287554868 hasAuthorship W4287554868A5060755739 @default.
- W4287554868 hasAuthorship W4287554868A5072391499 @default.
- W4287554868 hasBestOaLocation W42875548681 @default.
- W4287554868 hasConcept C112680207 @default.
- W4287554868 hasConcept C11413529 @default.
- W4287554868 hasConcept C126255220 @default.
- W4287554868 hasConcept C145446738 @default.
- W4287554868 hasConcept C154945302 @default.
- W4287554868 hasConcept C157972887 @default.
- W4287554868 hasConcept C17095337 @default.
- W4287554868 hasConcept C190502265 @default.
- W4287554868 hasConcept C2524010 @default.
- W4287554868 hasConcept C2776135515 @default.
- W4287554868 hasConcept C33923547 @default.
- W4287554868 hasConcept C41008148 @default.
- W4287554868 hasConcept C50644808 @default.
- W4287554868 hasConcept C81363708 @default.
- W4287554868 hasConceptScore W4287554868C112680207 @default.
- W4287554868 hasConceptScore W4287554868C11413529 @default.
- W4287554868 hasConceptScore W4287554868C126255220 @default.
- W4287554868 hasConceptScore W4287554868C145446738 @default.
- W4287554868 hasConceptScore W4287554868C154945302 @default.
- W4287554868 hasConceptScore W4287554868C157972887 @default.
- W4287554868 hasConceptScore W4287554868C17095337 @default.
- W4287554868 hasConceptScore W4287554868C190502265 @default.
- W4287554868 hasConceptScore W4287554868C2524010 @default.
- W4287554868 hasConceptScore W4287554868C2776135515 @default.
- W4287554868 hasConceptScore W4287554868C33923547 @default.
- W4287554868 hasConceptScore W4287554868C41008148 @default.
- W4287554868 hasConceptScore W4287554868C50644808 @default.
- W4287554868 hasConceptScore W4287554868C81363708 @default.
- W4287554868 hasLocation W42875548681 @default.
- W4287554868 hasOpenAccess W4287554868 @default.
- W4287554868 hasPrimaryLocation W42875548681 @default.
- W4287554868 hasRelatedWork W1995757014 @default.
- W4287554868 hasRelatedWork W2018944617 @default.
- W4287554868 hasRelatedWork W2081395119 @default.
- W4287554868 hasRelatedWork W2531890813 @default.
- W4287554868 hasRelatedWork W3003165794 @default.
- W4287554868 hasRelatedWork W3036599379 @default.
- W4287554868 hasRelatedWork W3203774161 @default.
- W4287554868 hasRelatedWork W4213315947 @default.
- W4287554868 hasRelatedWork W97914522 @default.
- W4287554868 hasRelatedWork W295059010 @default.
- W4287554868 isParatext "false" @default.
- W4287554868 isRetracted "false" @default.
- W4287554868 workType "article" @default.