FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964196211> ?p ?o ?g. }

• W2964196211 endingPage "1524" @default.
• W2964196211 startingPage "1516" @default.
• W2964196211 abstract "Several learning applications require solving high-dimensional regression problems where the relevant features belong to a small number of (overlapping) groups. For very large datasets and under standard sparsity constraints, hard thresholding methods have proven to be extremely efficient, but such methods require NP hard projections when dealing with overlapping groups. In this paper, we show that such NP-hard projections can not only be avoided by appealing to submodular optimization, but such methods come with strong theoretical guarantees even in the presence of poorly conditioned data (i.e. say when two features have correlation $geq 0.99$), which existing analyses cannot handle. These methods exhibit an interesting computation-accuracy trade-off and can be extended to significantly harder problems such as sparse overlapping groups. Experiments on both real and synthetic data validate our claims and demonstrate that the proposed methods are orders of magnitude faster than other greedy and convex relaxation techniques for learning with group-structured sparsity." @default.
• W2964196211 created "2019-07-30" @default.
• W2964196211 creator A5034432097 @default.
• W2964196211 creator A5063459703 @default.
• W2964196211 creator A5081428282 @default.
• W2964196211 date "2016-02-19" @default.
• W2964196211 modified "2023-09-24" @default.
• W2964196211 title "Structured Sparse Regression via Greedy Hard-Thresholding" @default.
• W2964196211 cites W1613006380 @default.
• W2964196211 cites W1680189815 @default.
• W2964196211 cites W1878643722 @default.
• W2964196211 cites W1970554427 @default.
• W2964196211 cites W1974160512 @default.
• W2964196211 cites W2015839735 @default.
• W2964196211 cites W2023697385 @default.
• W2964196211 cites W2118838680 @default.
• W2964196211 cites W2123023890 @default.
• W2964196211 cites W2130357047 @default.
• W2964196211 cites W2137565990 @default.
• W2964196211 cites W2158068051 @default.
• W2964196211 cites W2159268085 @default.
• W2964196211 cites W2163224677 @default.
• W2964196211 cites W2168893847 @default.
• W2964196211 cites W2208167373 @default.
• W2964196211 cites W2244252827 @default.
• W2964196211 cites W2963765119 @default.
• W2964196211 cites W33214042 @default.
• W2964196211 hasPublicationYear "2016" @default.
• W2964196211 type Work @default.
• W2964196211 sameAs 2964196211 @default.
• W2964196211 citedByCount "9" @default.
• W2964196211 countsByYear W29641962112017 @default.
• W2964196211 countsByYear W29641962112018 @default.
• W2964196211 countsByYear W29641962112019 @default.
• W2964196211 countsByYear W29641962112020 @default.
• W2964196211 countsByYear W29641962112021 @default.
• W2964196211 crossrefType "proceedings-article" @default.
• W2964196211 hasAuthorship W2964196211A5034432097 @default.
• W2964196211 hasAuthorship W2964196211A5063459703 @default.
• W2964196211 hasAuthorship W2964196211A5081428282 @default.
• W2964196211 hasConcept C105795698 @default.
• W2964196211 hasConcept C112680207 @default.
• W2964196211 hasConcept C11413529 @default.
• W2964196211 hasConcept C115961682 @default.
• W2964196211 hasConcept C126255220 @default.
• W2964196211 hasConcept C153180895 @default.
• W2964196211 hasConcept C154945302 @default.
• W2964196211 hasConcept C15744967 @default.
• W2964196211 hasConcept C178621042 @default.
• W2964196211 hasConcept C191178318 @default.
• W2964196211 hasConcept C2524010 @default.
• W2964196211 hasConcept C2776029896 @default.
• W2964196211 hasConcept C33923547 @default.
• W2964196211 hasConcept C41008148 @default.
• W2964196211 hasConcept C45374587 @default.
• W2964196211 hasConcept C51823790 @default.
• W2964196211 hasConcept C77805123 @default.
• W2964196211 hasConcept C83546350 @default.
• W2964196211 hasConceptScore W2964196211C105795698 @default.
• W2964196211 hasConceptScore W2964196211C112680207 @default.
• W2964196211 hasConceptScore W2964196211C11413529 @default.
• W2964196211 hasConceptScore W2964196211C115961682 @default.
• W2964196211 hasConceptScore W2964196211C126255220 @default.
• W2964196211 hasConceptScore W2964196211C153180895 @default.
• W2964196211 hasConceptScore W2964196211C154945302 @default.
• W2964196211 hasConceptScore W2964196211C15744967 @default.
• W2964196211 hasConceptScore W2964196211C178621042 @default.
• W2964196211 hasConceptScore W2964196211C191178318 @default.
• W2964196211 hasConceptScore W2964196211C2524010 @default.
• W2964196211 hasConceptScore W2964196211C2776029896 @default.
• W2964196211 hasConceptScore W2964196211C33923547 @default.
• W2964196211 hasConceptScore W2964196211C41008148 @default.
• W2964196211 hasConceptScore W2964196211C45374587 @default.
• W2964196211 hasConceptScore W2964196211C51823790 @default.
• W2964196211 hasConceptScore W2964196211C77805123 @default.
• W2964196211 hasConceptScore W2964196211C83546350 @default.
• W2964196211 hasLocation W29641962111 @default.
• W2964196211 hasOpenAccess W2964196211 @default.
• W2964196211 hasPrimaryLocation W29641962111 @default.
• W2964196211 hasRelatedWork W1523650926 @default.
• W2964196211 hasRelatedWork W1660214635 @default.
• W2964196211 hasRelatedWork W1838235479 @default.
• W2964196211 hasRelatedWork W2025079166 @default.
• W2964196211 hasRelatedWork W2054667730 @default.
• W2964196211 hasRelatedWork W2096725512 @default.
• W2964196211 hasRelatedWork W2105459687 @default.
• W2964196211 hasRelatedWork W2163224677 @default.
• W2964196211 hasRelatedWork W2250233051 @default.
• W2964196211 hasRelatedWork W2276322547 @default.
• W2964196211 hasRelatedWork W2289917018 @default.
• W2964196211 hasRelatedWork W2474269906 @default.
• W2964196211 hasRelatedWork W2552800523 @default.
• W2964196211 hasRelatedWork W2949447603 @default.
• W2964196211 hasRelatedWork W2952489486 @default.
• W2964196211 hasRelatedWork W2952677519 @default.
• W2964196211 hasRelatedWork W2963322354 @default.
• W2964196211 hasRelatedWork W2977663824 @default.
• W2964196211 hasRelatedWork W2996668298 @default.

• next