Matches in SemOpenAlex for { <https://semopenalex.org/work/W4293661513> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4293661513 abstract "This paper addresses the estimation of parameters of a Bayesian network from incomplete data. The task is usually tackled by running the Expectation-Maximization (EM) algorithm several times in order to obtain a high log-likelihood estimate. We argue that choosing the maximum log-likelihood estimate (as well as the maximum penalized log-likelihood and the maximum a posteriori estimate) has severe drawbacks, being affected both by overfitting and model uncertainty. Two ideas are discussed to overcome these issues: a maximum entropy approach and a Bayesian model averaging approach. Both ideas can be easily applied on top of EM, while the entropy idea can be also implemented in a more sophisticated way, through a dedicated non-linear solver. A vast set of experiments shows that these ideas produce significantly better estimates and inferences than the traditional and widely used maximum (penalized) log-likelihood and maximum a posteriori estimates. In particular, if EM is adopted as optimization engine, the model averaging approach is the best performing one; its performance is matched by the entropy approach when implemented using the non-linear solver. The results suggest that the applicability of these ideas is immediate (they are easy to implement and to integrate in currently available inference engines) and that they constitute a better way to learn Bayesian network parameters." @default.
- W4293661513 created "2022-08-31" @default.
- W4293661513 creator A5004994425 @default.
- W4293661513 creator A5047125862 @default.
- W4293661513 date "2011-10-12" @default.
- W4293661513 modified "2023-09-23" @default.
- W4293661513 title "Improving parameter learning of Bayesian nets from incomplete data" @default.
- W4293661513 doi "https://doi.org/10.48550/arxiv.1110.3239" @default.
- W4293661513 hasPublicationYear "2011" @default.
- W4293661513 type Work @default.
- W4293661513 citedByCount "0" @default.
- W4293661513 crossrefType "posted-content" @default.
- W4293661513 hasAuthorship W4293661513A5004994425 @default.
- W4293661513 hasAuthorship W4293661513A5047125862 @default.
- W4293661513 hasBestOaLocation W42936615131 @default.
- W4293661513 hasConcept C105795698 @default.
- W4293661513 hasConcept C107673813 @default.
- W4293661513 hasConcept C11413529 @default.
- W4293661513 hasConcept C119857082 @default.
- W4293661513 hasConcept C124101348 @default.
- W4293661513 hasConcept C126255220 @default.
- W4293661513 hasConcept C154945302 @default.
- W4293661513 hasConcept C160234255 @default.
- W4293661513 hasConcept C182081679 @default.
- W4293661513 hasConcept C199360897 @default.
- W4293661513 hasConcept C22019652 @default.
- W4293661513 hasConcept C2776214188 @default.
- W4293661513 hasConcept C2776330181 @default.
- W4293661513 hasConcept C2778770139 @default.
- W4293661513 hasConcept C33724603 @default.
- W4293661513 hasConcept C33923547 @default.
- W4293661513 hasConcept C41008148 @default.
- W4293661513 hasConcept C49781872 @default.
- W4293661513 hasConcept C50644808 @default.
- W4293661513 hasConcept C95923904 @default.
- W4293661513 hasConcept C9679016 @default.
- W4293661513 hasConcept C9810830 @default.
- W4293661513 hasConceptScore W4293661513C105795698 @default.
- W4293661513 hasConceptScore W4293661513C107673813 @default.
- W4293661513 hasConceptScore W4293661513C11413529 @default.
- W4293661513 hasConceptScore W4293661513C119857082 @default.
- W4293661513 hasConceptScore W4293661513C124101348 @default.
- W4293661513 hasConceptScore W4293661513C126255220 @default.
- W4293661513 hasConceptScore W4293661513C154945302 @default.
- W4293661513 hasConceptScore W4293661513C160234255 @default.
- W4293661513 hasConceptScore W4293661513C182081679 @default.
- W4293661513 hasConceptScore W4293661513C199360897 @default.
- W4293661513 hasConceptScore W4293661513C22019652 @default.
- W4293661513 hasConceptScore W4293661513C2776214188 @default.
- W4293661513 hasConceptScore W4293661513C2776330181 @default.
- W4293661513 hasConceptScore W4293661513C2778770139 @default.
- W4293661513 hasConceptScore W4293661513C33724603 @default.
- W4293661513 hasConceptScore W4293661513C33923547 @default.
- W4293661513 hasConceptScore W4293661513C41008148 @default.
- W4293661513 hasConceptScore W4293661513C49781872 @default.
- W4293661513 hasConceptScore W4293661513C50644808 @default.
- W4293661513 hasConceptScore W4293661513C95923904 @default.
- W4293661513 hasConceptScore W4293661513C9679016 @default.
- W4293661513 hasConceptScore W4293661513C9810830 @default.
- W4293661513 hasLocation W42936615131 @default.
- W4293661513 hasLocation W42936615132 @default.
- W4293661513 hasOpenAccess W4293661513 @default.
- W4293661513 hasPrimaryLocation W42936615131 @default.
- W4293661513 hasRelatedWork W1551198852 @default.
- W4293661513 hasRelatedWork W2101145897 @default.
- W4293661513 hasRelatedWork W2129130325 @default.
- W4293661513 hasRelatedWork W2183617661 @default.
- W4293661513 hasRelatedWork W2511279186 @default.
- W4293661513 hasRelatedWork W2963058055 @default.
- W4293661513 hasRelatedWork W3007584966 @default.
- W4293661513 hasRelatedWork W3029748970 @default.
- W4293661513 hasRelatedWork W4293661513 @default.
- W4293661513 hasRelatedWork W2126809555 @default.
- W4293661513 isParatext "false" @default.
- W4293661513 isRetracted "false" @default.
- W4293661513 workType "article" @default.