Matches in SemOpenAlex for { <https://semopenalex.org/work/W2953096735> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W2953096735 abstract "We consider structure discovery of undirected graphical models from observational data. Inferring likely structures from few examples is a complex task often requiring the formulation of priors and sophisticated inference procedures. Popular methods rely on estimating a penalized maximum likelihood of the precision matrix. However, in these approaches structure recovery is an indirect consequence of the data-fit term, the penalty can be difficult to adapt for domain-specific knowledge, and the inference is computationally demanding. By contrast, it may be easier to generate training samples of data that arise from graphs with the desired structure properties. We propose here to leverage this latter source of information as training data to learn a function, parametrized by a neural network that maps empirical covariance matrices to estimated graph structures. Learning this function brings two benefits: it implicitly models the desired structure or sparsity properties to form suitable priors, and it can be tailored to the specific problem of edge structure discovery, rather than maximizing data likelihood. Applying this framework, we find our learnable graph-discovery method trained on synthetic data generalizes well: identifying relevant edges in both synthetic and real data, completely unknown at training time. We find that on genetics, brain imaging, and simulation data we obtain performance generally superior to analytical methods." @default.
- W2953096735 created "2019-06-27" @default.
- W2953096735 creator A5025113992 @default.
- W2953096735 creator A5037751863 @default.
- W2953096735 creator A5074733625 @default.
- W2953096735 creator A5077783791 @default.
- W2953096735 date "2017-08-06" @default.
- W2953096735 modified "2023-10-04" @default.
- W2953096735 title "Learning to Discover Sparse Graphical Models" @default.
- W2953096735 cites W1528284822 @default.
- W2953096735 cites W1640882231 @default.
- W2953096735 cites W1906883763 @default.
- W2953096735 cites W1971600338 @default.
- W2953096735 cites W1989727964 @default.
- W2953096735 cites W2011650426 @default.
- W2953096735 cites W2062125287 @default.
- W2953096735 cites W2081825416 @default.
- W2953096735 cites W2095764512 @default.
- W2953096735 cites W2097581234 @default.
- W2953096735 cites W2099978770 @default.
- W2953096735 cites W2101234009 @default.
- W2953096735 cites W2112090702 @default.
- W2953096735 cites W2113517874 @default.
- W2953096735 cites W2118103795 @default.
- W2953096735 cites W2124227444 @default.
- W2953096735 cites W2132555912 @default.
- W2953096735 cites W2144499799 @default.
- W2953096735 cites W2148607939 @default.
- W2953096735 cites W2165009258 @default.
- W2953096735 cites W2167215970 @default.
- W2953096735 cites W2167868121 @default.
- W2953096735 cites W2173183968 @default.
- W2953096735 cites W2198196746 @default.
- W2953096735 cites W2286929393 @default.
- W2953096735 cites W2346728112 @default.
- W2953096735 cites W2507756961 @default.
- W2953096735 cites W2556967412 @default.
- W2953096735 cites W2949830731 @default.
- W2953096735 cites W2950898568 @default.
- W2953096735 cites W2951654389 @default.
- W2953096735 cites W2951779035 @default.
- W2953096735 cites W2962845550 @default.
- W2953096735 cites W2963330637 @default.
- W2953096735 cites W2964059111 @default.
- W2953096735 cites W2964121744 @default.
- W2953096735 cites W3103475671 @default.
- W2953096735 cites W637153065 @default.
- W2953096735 cites W2917998766 @default.
- W2953096735 hasPublicationYear "2017" @default.
- W2953096735 type Work @default.
- W2953096735 sameAs 2953096735 @default.
- W2953096735 citedByCount "2" @default.
- W2953096735 countsByYear W29530967352016 @default.
- W2953096735 countsByYear W29530967352017 @default.
- W2953096735 crossrefType "proceedings-article" @default.
- W2953096735 hasAuthorship W2953096735A5025113992 @default.
- W2953096735 hasAuthorship W2953096735A5037751863 @default.
- W2953096735 hasAuthorship W2953096735A5074733625 @default.
- W2953096735 hasAuthorship W2953096735A5077783791 @default.
- W2953096735 hasBestOaLocation W29530967351 @default.
- W2953096735 hasConcept C119857082 @default.
- W2953096735 hasConcept C154945302 @default.
- W2953096735 hasConcept C155846161 @default.
- W2953096735 hasConcept C41008148 @default.
- W2953096735 hasConceptScore W2953096735C119857082 @default.
- W2953096735 hasConceptScore W2953096735C154945302 @default.
- W2953096735 hasConceptScore W2953096735C155846161 @default.
- W2953096735 hasConceptScore W2953096735C41008148 @default.
- W2953096735 hasLocation W29530967351 @default.
- W2953096735 hasLocation W29530967352 @default.
- W2953096735 hasLocation W29530967353 @default.
- W2953096735 hasLocation W29530967354 @default.
- W2953096735 hasOpenAccess W2953096735 @default.
- W2953096735 hasPrimaryLocation W29530967351 @default.
- W2953096735 hasRelatedWork W2961085424 @default.
- W2953096735 hasRelatedWork W3046775127 @default.
- W2953096735 hasRelatedWork W3170094116 @default.
- W2953096735 hasRelatedWork W3209574120 @default.
- W2953096735 hasRelatedWork W4205958290 @default.
- W2953096735 hasRelatedWork W4285260836 @default.
- W2953096735 hasRelatedWork W4286629047 @default.
- W2953096735 hasRelatedWork W4306321456 @default.
- W2953096735 hasRelatedWork W4306674287 @default.
- W2953096735 hasRelatedWork W4224009465 @default.
- W2953096735 isParatext "false" @default.
- W2953096735 isRetracted "false" @default.
- W2953096735 magId "2953096735" @default.
- W2953096735 workType "article" @default.