Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950672178> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2950672178 abstract "We provide novel theoretical insights on structured prediction in the context of efficient convex surrogate loss minimization with consistency guarantees. For any task loss, we construct a convex surrogate that can be optimized via stochastic gradient descent and we prove tight bounds on the so-called calibration function relating the excess surrogate risk to the actual risk. In contrast to prior related work, we carefully monitor the effect of the exponential number of classes in the learning guarantees as well as on the optimization complexity. As an interesting consequence, we formalize the intuition that some task losses make learning harder than others, and that the classical 0-1 loss is ill-suited for general structured prediction." @default.
- W2950672178 created "2019-06-27" @default.
- W2950672178 creator A5001483226 @default.
- W2950672178 creator A5051076050 @default.
- W2950672178 creator A5059004658 @default.
- W2950672178 date "2017-12-04" @default.
- W2950672178 modified "2023-10-04" @default.
- W2950672178 title "On Structured Prediction Theory with Calibrated Convex Surrogate Losses" @default.
- W2950672178 cites W1480538416 @default.
- W2950672178 cites W1504372951 @default.
- W2950672178 cites W1515020792 @default.
- W2950672178 cites W1850680945 @default.
- W2950672178 cites W1978810924 @default.
- W2950672178 cites W1992208280 @default.
- W2950672178 cites W2008652694 @default.
- W2950672178 cites W2009570821 @default.
- W2950672178 cites W2023163512 @default.
- W2950672178 cites W2052448032 @default.
- W2950672178 cites W2097826433 @default.
- W2950672178 cites W2099119623 @default.
- W2950672178 cites W2101105183 @default.
- W2950672178 cites W2101767118 @default.
- W2950672178 cites W2105644991 @default.
- W2950672178 cites W2105842272 @default.
- W2950672178 cites W2108678517 @default.
- W2950672178 cites W2111161343 @default.
- W2950672178 cites W2113651538 @default.
- W2950672178 cites W2114028889 @default.
- W2950672178 cites W2115328410 @default.
- W2950672178 cites W2116444583 @default.
- W2950672178 cites W2136064009 @default.
- W2950672178 cites W2141789531 @default.
- W2950672178 cites W2143166796 @default.
- W2950672178 cites W2147880316 @default.
- W2950672178 cites W2151989798 @default.
- W2950672178 cites W2157791002 @default.
- W2950672178 cites W2158460069 @default.
- W2950672178 cites W2167247347 @default.
- W2950672178 cites W2169819436 @default.
- W2950672178 cites W2188209982 @default.
- W2950672178 cites W2212774062 @default.
- W2950672178 cites W2218302223 @default.
- W2950672178 cites W2403513640 @default.
- W2950672178 cites W2566190586 @default.
- W2950672178 cites W2582593119 @default.
- W2950672178 cites W2950754708 @default.
- W2950672178 cites W2963203807 @default.
- W2950672178 cites W2963343251 @default.
- W2950672178 cites W2963789586 @default.
- W2950672178 cites W581033782 @default.
- W2950672178 hasPublicationYear "2017" @default.
- W2950672178 type Work @default.
- W2950672178 sameAs 2950672178 @default.
- W2950672178 citedByCount "6" @default.
- W2950672178 countsByYear W29506721782017 @default.
- W2950672178 countsByYear W29506721782018 @default.
- W2950672178 countsByYear W29506721782021 @default.
- W2950672178 crossrefType "proceedings-article" @default.
- W2950672178 hasAuthorship W2950672178A5001483226 @default.
- W2950672178 hasAuthorship W2950672178A5051076050 @default.
- W2950672178 hasAuthorship W2950672178A5059004658 @default.
- W2950672178 hasBestOaLocation W29506721782 @default.
- W2950672178 hasConcept C112680207 @default.
- W2950672178 hasConcept C126255220 @default.
- W2950672178 hasConcept C2524010 @default.
- W2950672178 hasConcept C33923547 @default.
- W2950672178 hasConcept C41008148 @default.
- W2950672178 hasConceptScore W2950672178C112680207 @default.
- W2950672178 hasConceptScore W2950672178C126255220 @default.
- W2950672178 hasConceptScore W2950672178C2524010 @default.
- W2950672178 hasConceptScore W2950672178C33923547 @default.
- W2950672178 hasConceptScore W2950672178C41008148 @default.
- W2950672178 hasLocation W29506721781 @default.
- W2950672178 hasLocation W29506721782 @default.
- W2950672178 hasOpenAccess W2950672178 @default.
- W2950672178 hasPrimaryLocation W29506721781 @default.
- W2950672178 hasRelatedWork W2004154011 @default.
- W2950672178 hasRelatedWork W2018944617 @default.
- W2950672178 hasRelatedWork W2023469046 @default.
- W2950672178 hasRelatedWork W2518732165 @default.
- W2950672178 hasRelatedWork W2589063463 @default.
- W2950672178 hasRelatedWork W2747249330 @default.
- W2950672178 hasRelatedWork W2992250528 @default.
- W2950672178 hasRelatedWork W3046225436 @default.
- W2950672178 hasRelatedWork W4287703200 @default.
- W2950672178 hasRelatedWork W4297899479 @default.
- W2950672178 isParatext "false" @default.
- W2950672178 isRetracted "false" @default.
- W2950672178 magId "2950672178" @default.
- W2950672178 workType "article" @default.