Query Learning Strategies Using Boosting and Bagging

By Naoki Abe, Hiroshi Mamitsuka. 1998

How researchers see this paper

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

Citations in the corpus

Active Learning for Regression Based on Query by Committee (2007, methodology): “This approach is similar to query by bagging used in classification [3].”
Ensemble learning for data stream analysis: A survey (2017, methodology): “Bagging ++ [197] was developed as an improvement over Learn ++ by utilizing Bagging to construct new models from incoming chunks of data.”
Active Learning Literature Survey (2009, methodology): “For other model classes, such as discriminative or non-probabilistic models, Abe and Mamitsuka (1998) have proposed query-by-boosting and query-by-bagging , which employ the well-known ensemble learning methods boosting (Freund and Schapire, 1997) and bag-ging (Breiman, 1996) to construct…”
An Analysis of Active Learning Strategies for Sequence Labeling Tasks (2008, methodology): “In particular, we use the query-by-bagging approach (Abe and Mamitsuka, 1998) to learn a committee of CRFs.”
Exploiting Unlabeled Data in Content-Based Image Retrieval (2004, methodology): “The latter methods such as [1][15] generate a committee of several learners and select the unlabeled instances on which the committee members disagree the most.”
Reducing Annotation Effort for Quality Estimation via Active Learning (2013, methodology): “Since an SVR does not supply variance values for its predictions, we employ a technique known as query-by-bagging (Abe and Mamitsuka, 1998).”
Toward Optimal Active Learning through Sampling Estimation of Error Reduction (2001, methodology): “Abe and Mamitsuka (1998) use a bagging and boosting approach for maximizing the classifier accuracy on the test data.”
Active Semi-Supervision for Pairwise Constrained Clustering (2004, background): “, reduction of the version space size [16], reduction of uncertainty in predicted label [24], maximizing the margin on training data [1], finding high variance data points by density-weighted pool-based sampling [27], etc.”
A brief introduction to weakly supervised learning (2018, background): “The latter generates multiple learners and then queries the unlabeled instance on which the learners disagree to the most [1, 67].”
Semi-supervised learning by disagreement (2010, unknown): “Approaches of the former train a single learner and then query the unlabeled example on which the learner is least confident [45]; approaches of the latter generate multiple learners and then query the unlabeled example on which the learners disagree to the most [1] [63].”
A Survey of Active Learning Algorithms for Supervised Remote Sensing Image Classification (2011, methodology): “In the implementations of [25], bagging [29] is proposed to build the committee: first, training sets built on a draw with replacement of the original data are defined.”
Active Learning Methods for Remote Sensing Image Classification (2009, methodology): “In [36], methods based on boosting [37] and bagging [38] are described as adaptations of the query-by-”
Toward Optimal Active Learning through Monte Carlo Estimation of Error Reduction (2001, methodology): “In a return to the original QBC formulation, their approach is to estimate the reduction in classifier uncertainty by estimating the reduction in version space size as a function of querying instances.”
Batch-Mode Active-Learning Methods for the Interactive Classification of Remote Sensing Images (2011, methodology): “In [22], two query methods are proposed, which combine the idea of query by committee and that of boosting and bagging.”
Diverse ensembles for active learning (2004, methodology): “To tackle this issue, Abe and Mamitsuka (1998) proposed two variants of QBC, Query by Bagging and Query by Boosting, where Bagging and AdaBoost are used to construct the committees for sample selection.”
Outlier detection by active learning (2006, methodology): “[ 2 , 12]). The benefits of this approach are two-fold:”