Matches in SemOpenAlex for { <https://semopenalex.org/work/W3116293005> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W3116293005 abstract "We consider the differentiation of the value function for parametric optimization problems. Such problems are ubiquitous in Machine Learning applications such as structured support vector machines, matrix factorization and min-min or minimax problems in general. Existing approaches for computing the derivative rely on strong assumptions of the parametric function. Therefore, in several scenarios there is no theoretical evidence that a given algorithmic differentiation strategy computes the true gradient information of the value function. We leverage a well known result from convex duality theory to relax the conditions and to derive convergence rates of the derivative approximation for several classes of parametric optimization problems in Machine Learning. We demonstrate the versatility of our approach in several experiments, including non-smooth parametric functions. Even in settings where other approaches are applicable, our duality based strategy shows a favorable performance." @default.
- W3116293005 created "2021-01-05" @default.
- W3116293005 creator A5000059907 @default.
- W3116293005 creator A5082194697 @default.
- W3116293005 date "2020-12-27" @default.
- W3116293005 modified "2023-09-27" @default.
- W3116293005 title "Differentiating the Value Function by using Convex Duality" @default.
- W3116293005 hasPublicationYear "2020" @default.
- W3116293005 type Work @default.
- W3116293005 sameAs 3116293005 @default.
- W3116293005 citedByCount "0" @default.
- W3116293005 crossrefType "posted-content" @default.
- W3116293005 hasAuthorship W3116293005A5000059907 @default.
- W3116293005 hasAuthorship W3116293005A5082194697 @default.
- W3116293005 hasConcept C105795698 @default.
- W3116293005 hasConcept C112680207 @default.
- W3116293005 hasConcept C117251300 @default.
- W3116293005 hasConcept C118615104 @default.
- W3116293005 hasConcept C126255220 @default.
- W3116293005 hasConcept C137836250 @default.
- W3116293005 hasConcept C14036430 @default.
- W3116293005 hasConcept C145446738 @default.
- W3116293005 hasConcept C149728462 @default.
- W3116293005 hasConcept C153083717 @default.
- W3116293005 hasConcept C154945302 @default.
- W3116293005 hasConcept C157972887 @default.
- W3116293005 hasConcept C163863214 @default.
- W3116293005 hasConcept C2524010 @default.
- W3116293005 hasConcept C2778023678 @default.
- W3116293005 hasConcept C28826006 @default.
- W3116293005 hasConcept C33923547 @default.
- W3116293005 hasConcept C41008148 @default.
- W3116293005 hasConcept C5274546 @default.
- W3116293005 hasConcept C78458016 @default.
- W3116293005 hasConcept C86803240 @default.
- W3116293005 hasConceptScore W3116293005C105795698 @default.
- W3116293005 hasConceptScore W3116293005C112680207 @default.
- W3116293005 hasConceptScore W3116293005C117251300 @default.
- W3116293005 hasConceptScore W3116293005C118615104 @default.
- W3116293005 hasConceptScore W3116293005C126255220 @default.
- W3116293005 hasConceptScore W3116293005C137836250 @default.
- W3116293005 hasConceptScore W3116293005C14036430 @default.
- W3116293005 hasConceptScore W3116293005C145446738 @default.
- W3116293005 hasConceptScore W3116293005C149728462 @default.
- W3116293005 hasConceptScore W3116293005C153083717 @default.
- W3116293005 hasConceptScore W3116293005C154945302 @default.
- W3116293005 hasConceptScore W3116293005C157972887 @default.
- W3116293005 hasConceptScore W3116293005C163863214 @default.
- W3116293005 hasConceptScore W3116293005C2524010 @default.
- W3116293005 hasConceptScore W3116293005C2778023678 @default.
- W3116293005 hasConceptScore W3116293005C28826006 @default.
- W3116293005 hasConceptScore W3116293005C33923547 @default.
- W3116293005 hasConceptScore W3116293005C41008148 @default.
- W3116293005 hasConceptScore W3116293005C5274546 @default.
- W3116293005 hasConceptScore W3116293005C78458016 @default.
- W3116293005 hasConceptScore W3116293005C86803240 @default.
- W3116293005 hasLocation W31162930051 @default.
- W3116293005 hasOpenAccess W3116293005 @default.
- W3116293005 hasPrimaryLocation W31162930051 @default.
- W3116293005 hasRelatedWork W1786935204 @default.
- W3116293005 hasRelatedWork W2105180921 @default.
- W3116293005 hasRelatedWork W2166052342 @default.
- W3116293005 hasRelatedWork W2209633986 @default.
- W3116293005 hasRelatedWork W2243287592 @default.
- W3116293005 hasRelatedWork W2611385574 @default.
- W3116293005 hasRelatedWork W2769660485 @default.
- W3116293005 hasRelatedWork W2904593627 @default.
- W3116293005 hasRelatedWork W2907599741 @default.
- W3116293005 hasRelatedWork W2949581229 @default.
- W3116293005 hasRelatedWork W2950703405 @default.
- W3116293005 hasRelatedWork W2952554368 @default.
- W3116293005 hasRelatedWork W3014850082 @default.
- W3116293005 hasRelatedWork W3034736270 @default.
- W3116293005 hasRelatedWork W3080690820 @default.
- W3116293005 hasRelatedWork W3104368523 @default.
- W3116293005 hasRelatedWork W3122062812 @default.
- W3116293005 hasRelatedWork W3125338181 @default.
- W3116293005 hasRelatedWork W3159383416 @default.
- W3116293005 hasRelatedWork W633328766 @default.
- W3116293005 isParatext "false" @default.
- W3116293005 isRetracted "false" @default.
- W3116293005 magId "3116293005" @default.
- W3116293005 workType "article" @default.