Matches in SemOpenAlex for { <https://semopenalex.org/work/W3033249503> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3033249503 abstract "Designing appropriate variational regularization schemes is a crucial part of solving inverse problems, making them better-posed and guaranteeing that the solution of the associated optimization problem satisfies desirable properties. Recently, learning-based strategies have appeared to be very efficient for solving inverse problems, by learning direct inversion schemes or plug-and-play regularizers from available pairs of true states and observations. In this paper, we go a step further and design an end-to-end framework allowing to learn actual variational frameworks for inverse problems in such a supervised setting. The variational cost and the gradient-based solver are both stated as neural networks using automatic differentiation for the latter. We can jointly learn both components to minimize the data reconstruction error on the true states. This leads to a data-driven discovery of variational models. We consider an application to inverse problems with incomplete datasets (image inpainting and multivariate time series interpolation). We experimentally illustrate that this framework can lead to a significant gain in terms of reconstruction performance, including w.r.t. the direct minimization of the variational formulation derived from the known generative model." @default.
- W3033249503 created "2020-06-12" @default.
- W3033249503 creator A5045707906 @default.
- W3033249503 creator A5046811680 @default.
- W3033249503 creator A5075497720 @default.
- W3033249503 date "2020-06-05" @default.
- W3033249503 modified "2023-09-23" @default.
- W3033249503 title "Joint learning of variational representations and solvers for inverse problems with partially-observed data" @default.
- W3033249503 cites W1496099826 @default.
- W3033249503 cites W1510319310 @default.
- W3033249503 cites W1604521381 @default.
- W3033249503 cites W1915360731 @default.
- W3033249503 cites W1998680950 @default.
- W3033249503 cites W2005876975 @default.
- W3033249503 cites W2047071281 @default.
- W3033249503 cites W2065164181 @default.
- W3033249503 cites W2087416986 @default.
- W3033249503 cites W2101234009 @default.
- W3033249503 cites W2144144709 @default.
- W3033249503 cites W2146337213 @default.
- W3033249503 cites W2153663612 @default.
- W3033249503 cites W2552808051 @default.
- W3033249503 cites W2604885021 @default.
- W3033249503 cites W2782977076 @default.
- W3033249503 cites W2798365772 @default.
- W3033249503 cites W2804529069 @default.
- W3033249503 cites W2899283552 @default.
- W3033249503 cites W2902143762 @default.
- W3033249503 cites W2949440985 @default.
- W3033249503 cites W2950788055 @default.
- W3033249503 cites W2952020389 @default.
- W3033249503 cites W2954392630 @default.
- W3033249503 cites W2963226480 @default.
- W3033249503 cites W2963755523 @default.
- W3033249503 cites W2963775850 @default.
- W3033249503 cites W3015606043 @default.
- W3033249503 cites W3098900881 @default.
- W3033249503 cites W3101765447 @default.
- W3033249503 cites W3102025760 @default.
- W3033249503 cites W3102562067 @default.
- W3033249503 cites W1606472461 @default.
- W3033249503 hasPublicationYear "2020" @default.
- W3033249503 type Work @default.
- W3033249503 sameAs 3033249503 @default.
- W3033249503 citedByCount "2" @default.
- W3033249503 countsByYear W30332495032021 @default.
- W3033249503 crossrefType "posted-content" @default.
- W3033249503 hasAuthorship W3033249503A5045707906 @default.
- W3033249503 hasAuthorship W3033249503A5046811680 @default.
- W3033249503 hasAuthorship W3033249503A5075497720 @default.
- W3033249503 hasBestOaLocation W30332495031 @default.
- W3033249503 hasConcept C126255220 @default.
- W3033249503 hasConcept C127413603 @default.
- W3033249503 hasConcept C134306372 @default.
- W3033249503 hasConcept C135252773 @default.
- W3033249503 hasConcept C136119220 @default.
- W3033249503 hasConcept C154945302 @default.
- W3033249503 hasConcept C18555067 @default.
- W3033249503 hasConcept C202444582 @default.
- W3033249503 hasConcept C207467116 @default.
- W3033249503 hasConcept C2524010 @default.
- W3033249503 hasConcept C28826006 @default.
- W3033249503 hasConcept C33923547 @default.
- W3033249503 hasConcept C41008148 @default.
- W3033249503 hasConcept C66938386 @default.
- W3033249503 hasConceptScore W3033249503C126255220 @default.
- W3033249503 hasConceptScore W3033249503C127413603 @default.
- W3033249503 hasConceptScore W3033249503C134306372 @default.
- W3033249503 hasConceptScore W3033249503C135252773 @default.
- W3033249503 hasConceptScore W3033249503C136119220 @default.
- W3033249503 hasConceptScore W3033249503C154945302 @default.
- W3033249503 hasConceptScore W3033249503C18555067 @default.
- W3033249503 hasConceptScore W3033249503C202444582 @default.
- W3033249503 hasConceptScore W3033249503C207467116 @default.
- W3033249503 hasConceptScore W3033249503C2524010 @default.
- W3033249503 hasConceptScore W3033249503C28826006 @default.
- W3033249503 hasConceptScore W3033249503C33923547 @default.
- W3033249503 hasConceptScore W3033249503C41008148 @default.
- W3033249503 hasConceptScore W3033249503C66938386 @default.
- W3033249503 hasLocation W30332495031 @default.
- W3033249503 hasLocation W30332495032 @default.
- W3033249503 hasOpenAccess W3033249503 @default.
- W3033249503 hasPrimaryLocation W30332495031 @default.
- W3033249503 hasRelatedWork W1927677835 @default.
- W3033249503 hasRelatedWork W1965412793 @default.
- W3033249503 hasRelatedWork W1965977581 @default.
- W3033249503 hasRelatedWork W2078520445 @default.
- W3033249503 hasRelatedWork W2334039696 @default.
- W3033249503 hasRelatedWork W2339811963 @default.
- W3033249503 hasRelatedWork W2370581504 @default.
- W3033249503 hasRelatedWork W2963794791 @default.
- W3033249503 hasRelatedWork W297741754 @default.
- W3033249503 hasRelatedWork W1823792310 @default.
- W3033249503 isParatext "false" @default.
- W3033249503 isRetracted "false" @default.
- W3033249503 magId "3033249503" @default.
- W3033249503 workType "article" @default.