Matches in SemOpenAlex for { <https://semopenalex.org/work/W3102809253> ?p ?o ?g. }
- W3102809253 abstract "In this paper, we present the nonconvex exterior-point operator splitting (NExOS) algorithm, a novel linearly convergent first-order algorithm tailored for constrained nonconvex learning problems. We consider the problem of minimizing a convex cost function over a nonconvex constraint set, where projection onto the constraint set is single-valued around local minima. A wide range of nonconvex learning problems have this structure including (but not limited to) sparse and low-rank optimization problems. By exploiting the underlying geometry of the constraint set, NExOS finds a locally optimal point by solving a sequence of penalized problems with strictly decreasing penalty parameters. NExOS solves each penalized problem by applying an outer iteration operator splitting algorithm, which converges linearly to a local minimum of the corresponding penalized formulation under regularity conditions. Furthermore, the local minima of the penalized problems converge to a local minimum of the original problem as the penalty parameter goes to zero. We implement NExOS in the open-source Julia package NExOS.jl, which has been extensively tested on many instances from a wide variety of learning problems. We study several examples of well-known nonconvex learning problems, and we show that in spite of being general purpose, NExOS is able to compute high quality solutions very quickly and is competitive with specialized algorithms." @default.
- W3102809253 created "2020-11-23" @default.
- W3102809253 creator A5025454489 @default.
- W3102809253 creator A5081461094 @default.
- W3102809253 creator A5084053110 @default.
- W3102809253 date "2020-11-09" @default.
- W3102809253 modified "2023-09-27" @default.
- W3102809253 title "Exterior-point Operator Splitting for Nonconvex Learning" @default.
- W3102809253 cites W1523985187 @default.
- W3102809253 cites W1573659356 @default.
- W3102809253 cites W1972451512 @default.
- W3102809253 cites W1977901803 @default.
- W3102809253 cites W2014746566 @default.
- W3102809253 cites W2020167116 @default.
- W3102809253 cites W2043024229 @default.
- W3102809253 cites W205960364 @default.
- W3102809253 cites W2089294058 @default.
- W3102809253 cites W2097415784 @default.
- W3102809253 cites W2107861471 @default.
- W3102809253 cites W2117377454 @default.
- W3102809253 cites W2129520927 @default.
- W3102809253 cites W2142506503 @default.
- W3102809253 cites W2146130798 @default.
- W3102809253 cites W2146502635 @default.
- W3102809253 cites W2147656689 @default.
- W3102809253 cites W2164278908 @default.
- W3102809253 cites W2172099713 @default.
- W3102809253 cites W2184047134 @default.
- W3102809253 cites W2295046650 @default.
- W3102809253 cites W2296319761 @default.
- W3102809253 cites W2313629133 @default.
- W3102809253 cites W2336195973 @default.
- W3102809253 cites W2346438296 @default.
- W3102809253 cites W2496276084 @default.
- W3102809253 cites W2607116528 @default.
- W3102809253 cites W2611328865 @default.
- W3102809253 cites W2763081248 @default.
- W3102809253 cites W2913535645 @default.
- W3102809253 cites W2962853966 @default.
- W3102809253 cites W2963000588 @default.
- W3102809253 cites W2963213817 @default.
- W3102809253 cites W2963351303 @default.
- W3102809253 cites W2963627201 @default.
- W3102809253 cites W2963687412 @default.
- W3102809253 cites W2963750388 @default.
- W3102809253 cites W2963858139 @default.
- W3102809253 cites W2963942108 @default.
- W3102809253 cites W2964121744 @default.
- W3102809253 cites W2988569866 @default.
- W3102809253 cites W2993649961 @default.
- W3102809253 cites W3001585632 @default.
- W3102809253 cites W3011526999 @default.
- W3102809253 cites W3025638325 @default.
- W3102809253 cites W3103984481 @default.
- W3102809253 cites W3105393233 @default.
- W3102809253 cites W3105558219 @default.
- W3102809253 cites W3105772645 @default.
- W3102809253 cites W3124367822 @default.
- W3102809253 cites W416122618 @default.
- W3102809253 cites W2290452516 @default.
- W3102809253 hasPublicationYear "2020" @default.
- W3102809253 type Work @default.
- W3102809253 sameAs 3102809253 @default.
- W3102809253 citedByCount "0" @default.
- W3102809253 crossrefType "posted-content" @default.
- W3102809253 hasAuthorship W3102809253A5025454489 @default.
- W3102809253 hasAuthorship W3102809253A5081461094 @default.
- W3102809253 hasAuthorship W3102809253A5084053110 @default.
- W3102809253 hasConcept C104317684 @default.
- W3102809253 hasConcept C114614502 @default.
- W3102809253 hasConcept C126255220 @default.
- W3102809253 hasConcept C134306372 @default.
- W3102809253 hasConcept C137836250 @default.
- W3102809253 hasConcept C14036430 @default.
- W3102809253 hasConcept C158448853 @default.
- W3102809253 hasConcept C159985019 @default.
- W3102809253 hasConcept C164226766 @default.
- W3102809253 hasConcept C17020691 @default.
- W3102809253 hasConcept C177067428 @default.
- W3102809253 hasConcept C185592680 @default.
- W3102809253 hasConcept C186633575 @default.
- W3102809253 hasConcept C189237950 @default.
- W3102809253 hasConcept C192562407 @default.
- W3102809253 hasConcept C204323151 @default.
- W3102809253 hasConcept C2524010 @default.
- W3102809253 hasConcept C2776036281 @default.
- W3102809253 hasConcept C2778112365 @default.
- W3102809253 hasConcept C28719098 @default.
- W3102809253 hasConcept C33923547 @default.
- W3102809253 hasConcept C54355233 @default.
- W3102809253 hasConcept C55493867 @default.
- W3102809253 hasConcept C6180225 @default.
- W3102809253 hasConcept C78458016 @default.
- W3102809253 hasConcept C86339819 @default.
- W3102809253 hasConcept C86803240 @default.
- W3102809253 hasConceptScore W3102809253C104317684 @default.
- W3102809253 hasConceptScore W3102809253C114614502 @default.
- W3102809253 hasConceptScore W3102809253C126255220 @default.
- W3102809253 hasConceptScore W3102809253C134306372 @default.
- W3102809253 hasConceptScore W3102809253C137836250 @default.