FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950794910> ?p ?o ?g. }

• W2950794910 abstract "The iterations of many first-order algorithms, when applied to minimizing common regularized regression functions, often resemble neural network layers with pre-specified weights. This observation has prompted the development of learning-based approaches that purport to replace these iterations with enhanced surrogates forged as DNN models from available training data. For example, important NP-hard sparse estimation problems have recently benefitted from this genre of upgrade, with simple feedforward or recurrent networks ousting proximal gradient-based iterations. Analogously, this paper demonstrates that more powerful Bayesian algorithms for promoting sparsity, which rely on complex multi-loop majorization-minimization techniques, mirror the structure of more sophisticated long short-term memory (LSTM) networks, or alternative gated feedback networks previously designed for sequence prediction. As part of this development, we examine the parallels between latent variable trajectories operating across multiple time-scales during optimization, and the activations within deep network structures designed to adaptively model such characteristic sequences. The resulting insights lead to a novel sparse estimation system that, when granted training data, can estimate optimal solutions efficiently in regimes where other algorithms fail, including practical direction-of-arrival (DOA) and 3D geometry recovery problems. The underlying principles we expose are also suggestive of a learning process for a richer class of multi-loop algorithms in other domains." @default.
• W2950794910 created "2019-06-27" @default.
• W2950794910 creator A5053655489 @default.
• W2950794910 creator A5074745823 @default.
• W2950794910 creator A5085016531 @default.
• W2950794910 date "2017-06-09" @default.
• W2950794910 modified "2023-09-27" @default.
• W2950794910 title "From Bayesian Sparsity to Gated Recurrent Nets" @default.
• W2950794910 cites W1487520337 @default.
• W2950794910 cites W1648445109 @default.
• W2950794910 cites W1975089519 @default.
• W2950794910 cites W1978356057 @default.
• W2950794910 cites W1997413930 @default.
• W2950794910 cites W2012805954 @default.
• W2950794910 cites W2016980059 @default.
• W2950794910 cites W2028823365 @default.
• W2950794910 cites W2049633694 @default.
• W2950794910 cites W2057886711 @default.
• W2950794910 cites W2064675550 @default.
• W2950794910 cites W2100556411 @default.
• W2950794910 cites W2100649405 @default.
• W2950794910 cites W2107861471 @default.
• W2950794910 cites W2116611182 @default.
• W2950794910 cites W2118103795 @default.
• W2950794910 cites W2123023890 @default.
• W2950794910 cites W2127870457 @default.
• W2950794910 cites W2129131372 @default.
• W2950794910 cites W2135046866 @default.
• W2950794910 cites W2137012645 @default.
• W2950794910 cites W2142224912 @default.
• W2950794910 cites W2145096794 @default.
• W2950794910 cites W2149414429 @default.
• W2950794910 cites W2150962169 @default.
• W2950794910 cites W2296319761 @default.
• W2950794910 cites W2346728112 @default.
• W2950794910 cites W2427497464 @default.
• W2950794910 cites W2530325458 @default.
• W2950794910 cites W2619204584 @default.
• W2950794910 cites W2911546748 @default.
• W2950794910 cites W2950635152 @default.
• W2950794910 cites W2952276042 @default.
• W2950794910 cites W2953061907 @default.
• W2950794910 cites W2963322354 @default.
• W2950794910 cites W3098352114 @default.
• W2950794910 hasPublicationYear "2017" @default.
• W2950794910 type Work @default.
• W2950794910 sameAs 2950794910 @default.
• W2950794910 citedByCount "0" @default.
• W2950794910 crossrefType "posted-content" @default.
• W2950794910 hasAuthorship W2950794910A5053655489 @default.
• W2950794910 hasAuthorship W2950794910A5074745823 @default.
• W2950794910 hasAuthorship W2950794910A5085016531 @default.
• W2950794910 hasConcept C107673813 @default.
• W2950794910 hasConcept C111919701 @default.
• W2950794910 hasConcept C11413529 @default.
• W2950794910 hasConcept C119857082 @default.
• W2950794910 hasConcept C127413603 @default.
• W2950794910 hasConcept C133731056 @default.
• W2950794910 hasConcept C147168706 @default.
• W2950794910 hasConcept C147764199 @default.
• W2950794910 hasConcept C154945302 @default.
• W2950794910 hasConcept C199360897 @default.
• W2950794910 hasConcept C2778112365 @default.
• W2950794910 hasConcept C38858127 @default.
• W2950794910 hasConcept C41008148 @default.
• W2950794910 hasConcept C50644808 @default.
• W2950794910 hasConcept C51167844 @default.
• W2950794910 hasConcept C54355233 @default.
• W2950794910 hasConcept C86803240 @default.
• W2950794910 hasConcept C98045186 @default.
• W2950794910 hasConceptScore W2950794910C107673813 @default.
• W2950794910 hasConceptScore W2950794910C111919701 @default.
• W2950794910 hasConceptScore W2950794910C11413529 @default.
• W2950794910 hasConceptScore W2950794910C119857082 @default.
• W2950794910 hasConceptScore W2950794910C127413603 @default.
• W2950794910 hasConceptScore W2950794910C133731056 @default.
• W2950794910 hasConceptScore W2950794910C147168706 @default.
• W2950794910 hasConceptScore W2950794910C147764199 @default.
• W2950794910 hasConceptScore W2950794910C154945302 @default.
• W2950794910 hasConceptScore W2950794910C199360897 @default.
• W2950794910 hasConceptScore W2950794910C2778112365 @default.
• W2950794910 hasConceptScore W2950794910C38858127 @default.
• W2950794910 hasConceptScore W2950794910C41008148 @default.
• W2950794910 hasConceptScore W2950794910C50644808 @default.
• W2950794910 hasConceptScore W2950794910C51167844 @default.
• W2950794910 hasConceptScore W2950794910C54355233 @default.
• W2950794910 hasConceptScore W2950794910C86803240 @default.
• W2950794910 hasConceptScore W2950794910C98045186 @default.
• W2950794910 hasLocation W29507949101 @default.
• W2950794910 hasOpenAccess W2950794910 @default.
• W2950794910 hasPrimaryLocation W29507949101 @default.
• W2950794910 hasRelatedWork W1408639475 @default.
• W2950794910 hasRelatedWork W1831449718 @default.
• W2950794910 hasRelatedWork W2107306109 @default.
• W2950794910 hasRelatedWork W2162221686 @default.
• W2950794910 hasRelatedWork W2405778879 @default.
• W2950794910 hasRelatedWork W2416019854 @default.
• W2950794910 hasRelatedWork W2542136887 @default.
• W2950794910 hasRelatedWork W2557827285 @default.
• W2950794910 hasRelatedWork W2624699774 @default.

• next