MultiBoosting: A Technique for Combining Boosting and Wagging

By Geoffrey I. Webb. Machine-mediated learning, 2000

749 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Information Systems

View Paper: https://www.semanticscholar.org/paper/94eaee984b3017a3be3a386c3ae42b243494f62a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Information Systems

Citations in the corpus

Multistrategy ensemble learning: reducing error by combining ensemble learning techniques (2004, background): “While our specific technique (described in more detail by Zheng and Webb [21]) differs in minor details from these previous approaches, we have no reason to believe that, in general, its performance would differ substantially from the alternatives .”
Sentiment classification: The contribution of ensemble learning (2014, methodology): “We also employed the statistics used in [41] to compare two learning algorithms across all data sets, namely, the win/draw/loss record.”
Classification in the Presence of Label Noise: A Survey (2014, methodology): “A similar approach is proposed in [298]: the multiple boosting (MB) algorithm.”
Creating diversity in ensembles using artificial data (2005, methodology): “To evaluate the performance of Decorate we ran experiments on 15 representative data sets from the UCI repository (Blake and Merz, 1998) that were used in similar studies (Webb, 2000; Quinlan, 1996).”
A Survey on ensemble learning under the era of deep learning (2021, methodology): “Combining Bagging and Wagging, Webb (Webb 2000) proposed Multiboosting, which firstly uses Boosting to assign weights to the samples in the training data, and then applies Wagging to update the weights of the assigned samples.”
Protecting against evaluation overfitting in empirical reinforcement learning (2011, background): “Related issues have been noted in supervised machine learning [20], [21] and may explain why average accuracy across multiple data sets is rarely used to summarize performance in that community [21].”
Extremely randomized trees (2006, methodology): “So, it is very difficult to study randomization methods by trying to understand directly their effect on the misclassification error (see Bauer and Kohavi, 1999; Webb, 2000; Valentini and Dietterich, 2004, for empirical bias/variance analyses of ensemble methods).”
Deep forest (2017, result): “In contrast to previous studies about ensemble of ensembles, such as using Bagging as base learners for Boosting [58], gcForest uses the ensembles in the same grade together for feature re-representation.”
Statistical Comparisons of Classifiers over Multiple Data Sets (2006, background): “In words of Webb (2000), " it is debatable whether error rates in different domains are commensurable, and hence whether averaging error rates across domains is very mean- DEMŠAR ingful " .”
Out-of-bag estimation of the optimal sample size in bagging (2010, background): “Empirical studies have established that bagging is a simple and robust method that generally increases the accuracy of a single learner [1, 2, 3, 4, 5, 6, 7].”
Bagging gradient-boosted trees for high precision, low variance ranking models (2011, methodology): “There has been attempts for combining bias and variance reduction techniques for classiﬁcation [31, 29] and regression [5, 14, 26, 27] problems.”