Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210412346> ?p ?o ?g. }
- W3210412346 abstract "Ensembles are widely used in machine learning and, usually, provide state-of-the-art performance in many prediction tasks. From the very beginning, the diversity of an ensemble has been identified as a key factor for the superior performance of these models. But the exact role that diversity plays in ensemble models is poorly understood, specially in the context of neural networks. In this work, we combine and expand previously published results in a theoretically sound framework that describes the relationship between diversity and ensemble performance for a wide range of ensemble methods. More precisely, we provide sound answers to the following questions: how to measure diversity, how diversity relates to the generalization error of an ensemble, and how diversity is promoted by neural network ensemble algorithms. This analysis covers three widely used loss functions, namely, the squared loss, the cross-entropy loss, and the 0-1 loss; and two widely used model combination strategies, namely, model averaging and weighted majority vote. We empirically validate this theoretical analysis with neural network ensembles." @default.
- W3210412346 created "2021-11-08" @default.
- W3210412346 creator A5033099294 @default.
- W3210412346 creator A5053881061 @default.
- W3210412346 creator A5091366101 @default.
- W3210412346 date "2021-10-26" @default.
- W3210412346 modified "2023-09-27" @default.
- W3210412346 title "Diversity and Generalization in Neural Network Ensembles" @default.
- W3210412346 cites W1534477342 @default.
- W3210412346 cites W1539376383 @default.
- W3210412346 cites W1552624648 @default.
- W3210412346 cites W1580877334 @default.
- W3210412346 cites W1600469680 @default.
- W3210412346 cites W1696501374 @default.
- W3210412346 cites W1763872900 @default.
- W3210412346 cites W1903531865 @default.
- W3210412346 cites W1930161385 @default.
- W3210412346 cites W2014384147 @default.
- W3210412346 cites W2061119986 @default.
- W3210412346 cites W2067713319 @default.
- W3210412346 cites W2076118331 @default.
- W3210412346 cites W2091563134 @default.
- W3210412346 cites W2102605133 @default.
- W3210412346 cites W2103459159 @default.
- W3210412346 cites W2106979123 @default.
- W3210412346 cites W2112076978 @default.
- W3210412346 cites W2115629999 @default.
- W3210412346 cites W2116314036 @default.
- W3210412346 cites W2122892819 @default.
- W3210412346 cites W2124951716 @default.
- W3210412346 cites W2125555508 @default.
- W3210412346 cites W2128073546 @default.
- W3210412346 cites W2135293965 @default.
- W3210412346 cites W2143267104 @default.
- W3210412346 cites W2147800946 @default.
- W3210412346 cites W2163187564 @default.
- W3210412346 cites W2168022998 @default.
- W3210412346 cites W2182286517 @default.
- W3210412346 cites W2194775991 @default.
- W3210412346 cites W2402109343 @default.
- W3210412346 cites W2411932910 @default.
- W3210412346 cites W2467684189 @default.
- W3210412346 cites W2739485194 @default.
- W3210412346 cites W2911964244 @default.
- W3210412346 cites W2912070915 @default.
- W3210412346 cites W2912934387 @default.
- W3210412346 cites W2963238274 @default.
- W3210412346 cites W2963663068 @default.
- W3210412346 cites W2971130081 @default.
- W3210412346 cites W2992525328 @default.
- W3210412346 cites W2994797252 @default.
- W3210412346 cites W2996369242 @default.
- W3210412346 cites W2997129641 @default.
- W3210412346 cites W3095857158 @default.
- W3210412346 cites W3098341014 @default.
- W3210412346 cites W3106195168 @default.
- W3210412346 cites W3118608800 @default.
- W3210412346 cites W605727707 @default.
- W3210412346 hasPublicationYear "2021" @default.
- W3210412346 type Work @default.
- W3210412346 sameAs 3210412346 @default.
- W3210412346 citedByCount "0" @default.
- W3210412346 crossrefType "posted-content" @default.
- W3210412346 hasAuthorship W3210412346A5033099294 @default.
- W3210412346 hasAuthorship W3210412346A5053881061 @default.
- W3210412346 hasAuthorship W3210412346A5091366101 @default.
- W3210412346 hasConcept C106301342 @default.
- W3210412346 hasConcept C117765406 @default.
- W3210412346 hasConcept C119857082 @default.
- W3210412346 hasConcept C121332964 @default.
- W3210412346 hasConcept C134306372 @default.
- W3210412346 hasConcept C144024400 @default.
- W3210412346 hasConcept C151730666 @default.
- W3210412346 hasConcept C154945302 @default.
- W3210412346 hasConcept C167085575 @default.
- W3210412346 hasConcept C177148314 @default.
- W3210412346 hasConcept C19165224 @default.
- W3210412346 hasConcept C2779343474 @default.
- W3210412346 hasConcept C2781316041 @default.
- W3210412346 hasConcept C33923547 @default.
- W3210412346 hasConcept C41008148 @default.
- W3210412346 hasConcept C45942800 @default.
- W3210412346 hasConcept C50644808 @default.
- W3210412346 hasConcept C62520636 @default.
- W3210412346 hasConcept C86803240 @default.
- W3210412346 hasConceptScore W3210412346C106301342 @default.
- W3210412346 hasConceptScore W3210412346C117765406 @default.
- W3210412346 hasConceptScore W3210412346C119857082 @default.
- W3210412346 hasConceptScore W3210412346C121332964 @default.
- W3210412346 hasConceptScore W3210412346C134306372 @default.
- W3210412346 hasConceptScore W3210412346C144024400 @default.
- W3210412346 hasConceptScore W3210412346C151730666 @default.
- W3210412346 hasConceptScore W3210412346C154945302 @default.
- W3210412346 hasConceptScore W3210412346C167085575 @default.
- W3210412346 hasConceptScore W3210412346C177148314 @default.
- W3210412346 hasConceptScore W3210412346C19165224 @default.
- W3210412346 hasConceptScore W3210412346C2779343474 @default.
- W3210412346 hasConceptScore W3210412346C2781316041 @default.
- W3210412346 hasConceptScore W3210412346C33923547 @default.
- W3210412346 hasConceptScore W3210412346C41008148 @default.