Matches in SemOpenAlex for { <https://semopenalex.org/work/W323943631> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W323943631 abstract "approved: Thomas G. Dietterich Recent work has shown that AdaBoost can be viewed as an algorithm that maximizes the margin on the training data via functional gradient descent. Under this interpretation, the weight computed by AdaBoost, for each hypothesis generated, can be viewed as a step size parameter in a gradient descent search. Friedman has suggested that shrinking these step sizes could produce improved generalization and reduce overfitting. In a series of experiments, he showed that very small step sizes did indeed reduce overfitting and improve generalization for three variants of GradientBoost, his generic functional gradient descent algorithm. For this report, we tested whether reduced learning rates can also improve generalization in AdaBoost. We tested AdaBoost (applied to C4.5 decision trees) with reduced learning rates on 28 benchmark datasets. The results show that reduced learning rates provide no statistically significant improvement on these datasets. We conclude that reduced learning rates cannot be recommended for use with boosted decision trees on datasets similar to these benchmark datasets. Redacted for Privacy Shrunken Learning Rates Do Not Improve AdaBoost On Benchmark Datasets by Daniel L. K. Forrest" @default.
- W323943631 created "2016-06-24" @default.
- W323943631 creator A5017189216 @default.
- W323943631 date "2001-11-30" @default.
- W323943631 modified "2023-09-27" @default.
- W323943631 title "Shrunken learning rates do not improve AdaBoost on benchmark datasets" @default.
- W323943631 cites W1605688901 @default.
- W323943631 cites W1678356000 @default.
- W323943631 cites W1840338487 @default.
- W323943631 cites W1988790447 @default.
- W323943631 cites W2024046085 @default.
- W323943631 cites W2032210760 @default.
- W323943631 cites W2108263314 @default.
- W323943631 cites W2112076978 @default.
- W323943631 cites W2158144742 @default.
- W323943631 cites W2167277498 @default.
- W323943631 hasPublicationYear "2001" @default.
- W323943631 type Work @default.
- W323943631 sameAs 323943631 @default.
- W323943631 citedByCount "0" @default.
- W323943631 crossrefType "dissertation" @default.
- W323943631 hasAuthorship W323943631A5017189216 @default.
- W323943631 hasConcept C119857082 @default.
- W323943631 hasConcept C12267149 @default.
- W323943631 hasConcept C13280743 @default.
- W323943631 hasConcept C134306372 @default.
- W323943631 hasConcept C141404830 @default.
- W323943631 hasConcept C153180895 @default.
- W323943631 hasConcept C153258448 @default.
- W323943631 hasConcept C154945302 @default.
- W323943631 hasConcept C177148314 @default.
- W323943631 hasConcept C185798385 @default.
- W323943631 hasConcept C205649164 @default.
- W323943631 hasConcept C22019652 @default.
- W323943631 hasConcept C33923547 @default.
- W323943631 hasConcept C41008148 @default.
- W323943631 hasConcept C50644808 @default.
- W323943631 hasConcept C774472 @default.
- W323943631 hasConceptScore W323943631C119857082 @default.
- W323943631 hasConceptScore W323943631C12267149 @default.
- W323943631 hasConceptScore W323943631C13280743 @default.
- W323943631 hasConceptScore W323943631C134306372 @default.
- W323943631 hasConceptScore W323943631C141404830 @default.
- W323943631 hasConceptScore W323943631C153180895 @default.
- W323943631 hasConceptScore W323943631C153258448 @default.
- W323943631 hasConceptScore W323943631C154945302 @default.
- W323943631 hasConceptScore W323943631C177148314 @default.
- W323943631 hasConceptScore W323943631C185798385 @default.
- W323943631 hasConceptScore W323943631C205649164 @default.
- W323943631 hasConceptScore W323943631C22019652 @default.
- W323943631 hasConceptScore W323943631C33923547 @default.
- W323943631 hasConceptScore W323943631C41008148 @default.
- W323943631 hasConceptScore W323943631C50644808 @default.
- W323943631 hasConceptScore W323943631C774472 @default.
- W323943631 hasLocation W3239436311 @default.
- W323943631 hasOpenAccess W323943631 @default.
- W323943631 hasPrimaryLocation W3239436311 @default.
- W323943631 hasRelatedWork W1522999595 @default.
- W323943631 hasRelatedWork W1569376970 @default.
- W323943631 hasRelatedWork W1963414467 @default.
- W323943631 hasRelatedWork W2104634417 @default.
- W323943631 hasRelatedWork W2293512291 @default.
- W323943631 hasRelatedWork W2530727159 @default.
- W323943631 hasRelatedWork W2588588350 @default.
- W323943631 hasRelatedWork W2783489654 @default.
- W323943631 hasRelatedWork W28154616 @default.
- W323943631 hasRelatedWork W2898821662 @default.
- W323943631 hasRelatedWork W2900892731 @default.
- W323943631 hasRelatedWork W2916383701 @default.
- W323943631 hasRelatedWork W2951434536 @default.
- W323943631 hasRelatedWork W2964161291 @default.
- W323943631 hasRelatedWork W2965376962 @default.
- W323943631 hasRelatedWork W2980600608 @default.
- W323943631 hasRelatedWork W3010114384 @default.
- W323943631 hasRelatedWork W3103064266 @default.
- W323943631 hasRelatedWork W3173320634 @default.
- W323943631 hasRelatedWork W3174302393 @default.
- W323943631 isParatext "false" @default.
- W323943631 isRetracted "false" @default.
- W323943631 magId "323943631" @default.
- W323943631 workType "dissertation" @default.