Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891778781> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W2891778781 endingPage "6821" @default.
- W2891778781 startingPage "6810" @default.
- W2891778781 abstract "Many structured prediction problems admit a natural loss function for evaluation such as the edit-distance or $n$-gram loss. However, existing learning algorithms are typically designed to optimize alternative objectives such as the cross-entropy. This is because a na{i}ve implementation of the natural loss functions often results in intractable gradient computations. In this paper, we design efficient gradient computation algorithms for two broad families of structured prediction loss functions: rational and tropical losses. These families include as special cases the $n$-gram loss, the edit-distance loss, and many other loss functions commonly used in natural language processing and computational biology tasks that are based on sequence similarity measures. Our algorithms make use of weighted automata and graph operations over appropriate semirings to design efficient solutions. They facilitate efficient gradient computation and hence enable one to train learning models such as neural networks with complex structured losses." @default.
- W2891778781 created "2018-09-27" @default.
- W2891778781 creator A5013243115 @default.
- W2891778781 creator A5047544753 @default.
- W2891778781 creator A5058849006 @default.
- W2891778781 creator A5065866283 @default.
- W2891778781 creator A5090764070 @default.
- W2891778781 date "2018-01-01" @default.
- W2891778781 modified "2023-10-18" @default.
- W2891778781 title "Efficient Gradient Computation for Structured Output Learning with Rational and Tropical Losses" @default.
- W2891778781 hasPublicationYear "2018" @default.
- W2891778781 type Work @default.
- W2891778781 sameAs 2891778781 @default.
- W2891778781 citedByCount "1" @default.
- W2891778781 countsByYear W28917787812021 @default.
- W2891778781 crossrefType "proceedings-article" @default.
- W2891778781 hasAuthorship W2891778781A5013243115 @default.
- W2891778781 hasAuthorship W2891778781A5047544753 @default.
- W2891778781 hasAuthorship W2891778781A5058849006 @default.
- W2891778781 hasAuthorship W2891778781A5065866283 @default.
- W2891778781 hasAuthorship W2891778781A5090764070 @default.
- W2891778781 hasConcept C106301342 @default.
- W2891778781 hasConcept C11413529 @default.
- W2891778781 hasConcept C119857082 @default.
- W2891778781 hasConcept C121332964 @default.
- W2891778781 hasConcept C132525143 @default.
- W2891778781 hasConcept C154945302 @default.
- W2891778781 hasConcept C41008148 @default.
- W2891778781 hasConcept C44359876 @default.
- W2891778781 hasConcept C45374587 @default.
- W2891778781 hasConcept C62520636 @default.
- W2891778781 hasConcept C80444323 @default.
- W2891778781 hasConceptScore W2891778781C106301342 @default.
- W2891778781 hasConceptScore W2891778781C11413529 @default.
- W2891778781 hasConceptScore W2891778781C119857082 @default.
- W2891778781 hasConceptScore W2891778781C121332964 @default.
- W2891778781 hasConceptScore W2891778781C132525143 @default.
- W2891778781 hasConceptScore W2891778781C154945302 @default.
- W2891778781 hasConceptScore W2891778781C41008148 @default.
- W2891778781 hasConceptScore W2891778781C44359876 @default.
- W2891778781 hasConceptScore W2891778781C45374587 @default.
- W2891778781 hasConceptScore W2891778781C62520636 @default.
- W2891778781 hasConceptScore W2891778781C80444323 @default.
- W2891778781 hasLocation W28917787811 @default.
- W2891778781 hasOpenAccess W2891778781 @default.
- W2891778781 hasPrimaryLocation W28917787811 @default.
- W2891778781 hasRelatedWork W2015342278 @default.
- W2891778781 hasRelatedWork W2028818097 @default.
- W2891778781 hasRelatedWork W2054667730 @default.
- W2891778781 hasRelatedWork W2091428542 @default.
- W2891778781 hasRelatedWork W2097707447 @default.
- W2891778781 hasRelatedWork W2105459687 @default.
- W2891778781 hasRelatedWork W2107108030 @default.
- W2891778781 hasRelatedWork W2248157934 @default.
- W2891778781 hasRelatedWork W2562068195 @default.
- W2891778781 hasRelatedWork W268427563 @default.
- W2891778781 hasRelatedWork W2766307072 @default.
- W2891778781 hasRelatedWork W2775077303 @default.
- W2891778781 hasRelatedWork W2963638618 @default.
- W2891778781 hasRelatedWork W2964056091 @default.
- W2891778781 hasRelatedWork W2990874390 @default.
- W2891778781 hasRelatedWork W3092532932 @default.
- W2891778781 hasRelatedWork W3134556739 @default.
- W2891778781 hasRelatedWork W3165174865 @default.
- W2891778781 hasRelatedWork W3212121437 @default.
- W2891778781 hasRelatedWork W2302853122 @default.
- W2891778781 hasVolume "31" @default.
- W2891778781 isParatext "false" @default.
- W2891778781 isRetracted "false" @default.
- W2891778781 magId "2891778781" @default.
- W2891778781 workType "article" @default.