Matches in SemOpenAlex for { <https://semopenalex.org/work/W3096802140> ?p ?o ?g. }
- W3096802140 abstract "Bregman divergences generalize measures such as the squared Euclidean distance and the KL divergence, and arise throughout many areas of machine learning. In this paper, we focus on the problem of approximating an arbitrary Bregman divergence from supervision, and we provide a well-principled approach to analyzing such approximations. We develop a formulation and algorithm for learning arbitrary Bregman divergences based on approximating their underlying convex generating function via a piecewise linear function. We provide theoretical approximation bounds using our parameterization and show that the generalization error $O_p(m^{-1/2})$ for metric learning using our framework matches the known generalization error in the strictly less general Mahalanobis metric learning setting. We further demonstrate empirically that our method performs well in comparison to existing metric learning methods, particularly for clustering and ranking problems." @default.
- W3096802140 created "2020-11-09" @default.
- W3096802140 creator A5007855657 @default.
- W3096802140 creator A5048704387 @default.
- W3096802140 creator A5049134079 @default.
- W3096802140 creator A5058386500 @default.
- W3096802140 creator A5091179204 @default.
- W3096802140 date "2019-05-28" @default.
- W3096802140 modified "2023-10-17" @default.
- W3096802140 title "Learning to Approximate a Bregman Divergence" @default.
- W3096802140 cites W1490654344 @default.
- W3096802140 cites W1542886316 @default.
- W3096802140 cites W1584092455 @default.
- W3096802140 cites W1973264045 @default.
- W3096802140 cites W1973435495 @default.
- W3096802140 cites W2062434421 @default.
- W3096802140 cites W2096765209 @default.
- W3096802140 cites W2103183297 @default.
- W3096802140 cites W2105280352 @default.
- W3096802140 cites W2106053110 @default.
- W3096802140 cites W2115781554 @default.
- W3096802140 cites W2117154949 @default.
- W3096802140 cites W2118393783 @default.
- W3096802140 cites W2118979615 @default.
- W3096802140 cites W2121949863 @default.
- W3096802140 cites W2134089414 @default.
- W3096802140 cites W213438331 @default.
- W3096802140 cites W2135442311 @default.
- W3096802140 cites W2136691247 @default.
- W3096802140 cites W2138021189 @default.
- W3096802140 cites W2140376886 @default.
- W3096802140 cites W2144935315 @default.
- W3096802140 cites W2147987877 @default.
- W3096802140 cites W2150038733 @default.
- W3096802140 cites W2158139921 @default.
- W3096802140 cites W2169495281 @default.
- W3096802140 cites W2468075526 @default.
- W3096802140 cites W2803119945 @default.
- W3096802140 cites W2963775347 @default.
- W3096802140 cites W3035690202 @default.
- W3096802140 cites W607505555 @default.
- W3096802140 hasPublicationYear "2019" @default.
- W3096802140 type Work @default.
- W3096802140 sameAs 3096802140 @default.
- W3096802140 citedByCount "2" @default.
- W3096802140 countsByYear W30968021402020 @default.
- W3096802140 countsByYear W30968021402021 @default.
- W3096802140 crossrefType "posted-content" @default.
- W3096802140 hasAuthorship W3096802140A5007855657 @default.
- W3096802140 hasAuthorship W3096802140A5048704387 @default.
- W3096802140 hasAuthorship W3096802140A5049134079 @default.
- W3096802140 hasAuthorship W3096802140A5058386500 @default.
- W3096802140 hasAuthorship W3096802140A5091179204 @default.
- W3096802140 hasConcept C112680207 @default.
- W3096802140 hasConcept C120174047 @default.
- W3096802140 hasConcept C120665830 @default.
- W3096802140 hasConcept C121332964 @default.
- W3096802140 hasConcept C126255220 @default.
- W3096802140 hasConcept C129782007 @default.
- W3096802140 hasConcept C134306372 @default.
- W3096802140 hasConcept C138885662 @default.
- W3096802140 hasConcept C14036430 @default.
- W3096802140 hasConcept C149073432 @default.
- W3096802140 hasConcept C154945302 @default.
- W3096802140 hasConcept C162324750 @default.
- W3096802140 hasConcept C164660894 @default.
- W3096802140 hasConcept C176217482 @default.
- W3096802140 hasConcept C177148314 @default.
- W3096802140 hasConcept C189430467 @default.
- W3096802140 hasConcept C1921717 @default.
- W3096802140 hasConcept C192209626 @default.
- W3096802140 hasConcept C207390915 @default.
- W3096802140 hasConcept C21547014 @default.
- W3096802140 hasConcept C2524010 @default.
- W3096802140 hasConcept C28826006 @default.
- W3096802140 hasConcept C33923547 @default.
- W3096802140 hasConcept C41008148 @default.
- W3096802140 hasConcept C41895202 @default.
- W3096802140 hasConcept C73555534 @default.
- W3096802140 hasConcept C78458016 @default.
- W3096802140 hasConcept C86803240 @default.
- W3096802140 hasConceptScore W3096802140C112680207 @default.
- W3096802140 hasConceptScore W3096802140C120174047 @default.
- W3096802140 hasConceptScore W3096802140C120665830 @default.
- W3096802140 hasConceptScore W3096802140C121332964 @default.
- W3096802140 hasConceptScore W3096802140C126255220 @default.
- W3096802140 hasConceptScore W3096802140C129782007 @default.
- W3096802140 hasConceptScore W3096802140C134306372 @default.
- W3096802140 hasConceptScore W3096802140C138885662 @default.
- W3096802140 hasConceptScore W3096802140C14036430 @default.
- W3096802140 hasConceptScore W3096802140C149073432 @default.
- W3096802140 hasConceptScore W3096802140C154945302 @default.
- W3096802140 hasConceptScore W3096802140C162324750 @default.
- W3096802140 hasConceptScore W3096802140C164660894 @default.
- W3096802140 hasConceptScore W3096802140C176217482 @default.
- W3096802140 hasConceptScore W3096802140C177148314 @default.
- W3096802140 hasConceptScore W3096802140C189430467 @default.
- W3096802140 hasConceptScore W3096802140C1921717 @default.
- W3096802140 hasConceptScore W3096802140C192209626 @default.
- W3096802140 hasConceptScore W3096802140C207390915 @default.