Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320559890> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W4320559890 abstract "Bilevel optimization enjoys a wide range of applications in hyper-parameter optimization, meta-learning and reinforcement learning. However, bilevel optimization problems are difficult to solve. Recent progress on scalable bilevel algorithms mainly focuses on bilevel optimization problems where the lower-level objective is either strongly convex or unconstrained. In this work, we tackle the bilevel problem through the lens of the penalty method. We show that under certain conditions, the penalty reformulation recovers the solutions of the original bilevel problem. Further, we propose the penalty-based bilevel gradient descent (PBGD) algorithm and establish its finite-time convergence for the constrained bilevel problem without lower-level strong convexity. Experiments showcase the efficiency of the proposed PBGD algorithm." @default.
- W4320559890 created "2023-02-14" @default.
- W4320559890 creator A5026487799 @default.
- W4320559890 creator A5090323845 @default.
- W4320559890 date "2023-02-10" @default.
- W4320559890 modified "2023-09-25" @default.
- W4320559890 title "On Penalty-based Bilevel Gradient Descent Method" @default.
- W4320559890 doi "https://doi.org/10.48550/arxiv.2302.05185" @default.
- W4320559890 hasPublicationYear "2023" @default.
- W4320559890 type Work @default.
- W4320559890 citedByCount "0" @default.
- W4320559890 crossrefType "posted-content" @default.
- W4320559890 hasAuthorship W4320559890A5026487799 @default.
- W4320559890 hasAuthorship W4320559890A5090323845 @default.
- W4320559890 hasBestOaLocation W43205598901 @default.
- W4320559890 hasConcept C106159729 @default.
- W4320559890 hasConcept C126255220 @default.
- W4320559890 hasConcept C137836250 @default.
- W4320559890 hasConcept C153258448 @default.
- W4320559890 hasConcept C154945302 @default.
- W4320559890 hasConcept C162324750 @default.
- W4320559890 hasConcept C2777303404 @default.
- W4320559890 hasConcept C3309286 @default.
- W4320559890 hasConcept C33923547 @default.
- W4320559890 hasConcept C41008148 @default.
- W4320559890 hasConcept C50522688 @default.
- W4320559890 hasConcept C50644808 @default.
- W4320559890 hasConcept C6180225 @default.
- W4320559890 hasConcept C72134830 @default.
- W4320559890 hasConceptScore W4320559890C106159729 @default.
- W4320559890 hasConceptScore W4320559890C126255220 @default.
- W4320559890 hasConceptScore W4320559890C137836250 @default.
- W4320559890 hasConceptScore W4320559890C153258448 @default.
- W4320559890 hasConceptScore W4320559890C154945302 @default.
- W4320559890 hasConceptScore W4320559890C162324750 @default.
- W4320559890 hasConceptScore W4320559890C2777303404 @default.
- W4320559890 hasConceptScore W4320559890C3309286 @default.
- W4320559890 hasConceptScore W4320559890C33923547 @default.
- W4320559890 hasConceptScore W4320559890C41008148 @default.
- W4320559890 hasConceptScore W4320559890C50522688 @default.
- W4320559890 hasConceptScore W4320559890C50644808 @default.
- W4320559890 hasConceptScore W4320559890C6180225 @default.
- W4320559890 hasConceptScore W4320559890C72134830 @default.
- W4320559890 hasLocation W43205598901 @default.
- W4320559890 hasOpenAccess W4320559890 @default.
- W4320559890 hasPrimaryLocation W43205598901 @default.
- W4320559890 hasRelatedWork W1988507758 @default.
- W4320559890 hasRelatedWork W1994921083 @default.
- W4320559890 hasRelatedWork W2013038594 @default.
- W4320559890 hasRelatedWork W2119826439 @default.
- W4320559890 hasRelatedWork W2298416129 @default.
- W4320559890 hasRelatedWork W2313503396 @default.
- W4320559890 hasRelatedWork W2783384756 @default.
- W4320559890 hasRelatedWork W2995990939 @default.
- W4320559890 hasRelatedWork W3148242583 @default.
- W4320559890 hasRelatedWork W4312397237 @default.
- W4320559890 isParatext "false" @default.
- W4320559890 isRetracted "false" @default.
- W4320559890 workType "article" @default.