Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204364042> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W3204364042 abstract "Deep learning models have been very successful in the application of machine learning methods,often out-performing classical statistical models such as linear regression models orgeneralized linear models. On the other hand, deep learning models are often criticizedfor not being explainable nor allowing for variable selection. There are two different waysof dealing with this problem, either we use post-hoc model interpretability methods or wedesign specific deep learning architectures that allow for (more) easy interpretation and explanation. This paper builds on our previous work on the LocalGLMnet architecture thatgives an interpretable deep learning architecture. In the present paper, we show how groupLASSO regularization (and other regularization schemes) can be implemented within theLocalGLMnet architecture so that we receive feature sparsity for variable selection. Webenchmark our approach with the recently developed LassoNet of Lemhadri et al." @default.
- W3204364042 created "2021-10-11" @default.
- W3204364042 creator A5004533209 @default.
- W3204364042 creator A5048936137 @default.
- W3204364042 date "2021-01-01" @default.
- W3204364042 modified "2023-10-14" @default.
- W3204364042 title "LASSO Regularization within the LocalGLMnet Architecture" @default.
- W3204364042 cites W1523985187 @default.
- W3204364042 cites W1678356000 @default.
- W3204364042 cites W2025720061 @default.
- W3204364042 cites W2029469881 @default.
- W3204364042 cites W2073300221 @default.
- W3204364042 cites W2075965721 @default.
- W3204364042 cites W2122825543 @default.
- W3204364042 cites W2138019504 @default.
- W3204364042 cites W2140514146 @default.
- W3204364042 cites W3035517615 @default.
- W3204364042 cites W3206660074 @default.
- W3204364042 cites W4234698323 @default.
- W3204364042 doi "https://doi.org/10.2139/ssrn.3927187" @default.
- W3204364042 hasPublicationYear "2021" @default.
- W3204364042 type Work @default.
- W3204364042 sameAs 3204364042 @default.
- W3204364042 citedByCount "1" @default.
- W3204364042 countsByYear W32043640422022 @default.
- W3204364042 crossrefType "journal-article" @default.
- W3204364042 hasAuthorship W3204364042A5004533209 @default.
- W3204364042 hasAuthorship W3204364042A5048936137 @default.
- W3204364042 hasConcept C123657996 @default.
- W3204364042 hasConcept C154945302 @default.
- W3204364042 hasConcept C166957645 @default.
- W3204364042 hasConcept C199360897 @default.
- W3204364042 hasConcept C205649164 @default.
- W3204364042 hasConcept C2776135515 @default.
- W3204364042 hasConcept C33923547 @default.
- W3204364042 hasConcept C37616216 @default.
- W3204364042 hasConcept C41008148 @default.
- W3204364042 hasConceptScore W3204364042C123657996 @default.
- W3204364042 hasConceptScore W3204364042C154945302 @default.
- W3204364042 hasConceptScore W3204364042C166957645 @default.
- W3204364042 hasConceptScore W3204364042C199360897 @default.
- W3204364042 hasConceptScore W3204364042C205649164 @default.
- W3204364042 hasConceptScore W3204364042C2776135515 @default.
- W3204364042 hasConceptScore W3204364042C33923547 @default.
- W3204364042 hasConceptScore W3204364042C37616216 @default.
- W3204364042 hasConceptScore W3204364042C41008148 @default.
- W3204364042 hasLocation W32043640421 @default.
- W3204364042 hasOpenAccess W3204364042 @default.
- W3204364042 hasPrimaryLocation W32043640421 @default.
- W3204364042 hasRelatedWork W2130579308 @default.
- W3204364042 hasRelatedWork W2148971169 @default.
- W3204364042 hasRelatedWork W2158914562 @default.
- W3204364042 hasRelatedWork W2382623646 @default.
- W3204364042 hasRelatedWork W2762577149 @default.
- W3204364042 hasRelatedWork W2901346193 @default.
- W3204364042 hasRelatedWork W2972850838 @default.
- W3204364042 hasRelatedWork W3107474891 @default.
- W3204364042 hasRelatedWork W396164270 @default.
- W3204364042 hasRelatedWork W4368274933 @default.
- W3204364042 isParatext "false" @default.
- W3204364042 isRetracted "false" @default.
- W3204364042 magId "3204364042" @default.
- W3204364042 workType "article" @default.