paper-with-me

Papers

Optimally Combining Classifiers Using Unlabeled Data

2015-03-05 · Akshay Balsubramani, Yoav Freund

We develop a worst-case analysis of aggregation of classifier ensembles for binary classification. The task of predicting to minimize error is formulated as a game played over a given set of unlabeled data (a transductive setting), where prior label information is encoded as constraints on the game. The minimax solution of this game identifies cases where a weighted combination of the classifiers can perform significantly better than any single classifier.

📄 PDF Abstract BibTeX arXiv:1503.01811

Code (1)

aikanor/marvin

Tasks

Binary ClassificationGeneral Classification

Similar Papers 제목 키워드 기반

Optimally Combining Classifiers for Semi-Supervised Learning

2020-06-07 · Zhiguo Wang, Liusha Yang, Feng Yin, Ke Lin 외

This paper considers semi-supervised learning for tabular data. It is widely known that Xgboost based on tree model works well on the heterogeneous features while transductive support vector machine can exploit the low d…

Diversity

Estimating Accuracy from Unlabeled Data: A Probabilistic Logic Approach

2017-05-19 · NeurIPS 2017 12 · Emmanouil A. Platanios, Hoifung Poon, Tom M. Mitchell, Eric Horvitz

We propose an efficient method to estimate the accuracy of classifiers using only unlabeled data. We consider a setting with multiple classification problems where the target classes may be tied together through logical …

Sample-Specific Late Fusion for Visual Category Recognition

2013-06-01 · CVPR 2013 6 · Dong Liu, Kuan-Ting Lai, Guangnan Ye, Ming-Syan Chen 외

Late fusion addresses the problem of combining the prediction scores of multiple classifiers, in which each score is predicted by a classifier trained with a specific feature. However, the existing methods generally use …

Agreement Rate Initialized Maximum Likelihood Estimator for Ensemble Classifier Aggregation and Its Application in Brain-Computer Interface

2018-05-12 · Dongrui Wu, Vernon J. Lawhern, Stephen Gordon, Brent J. Lance 외

Ensemble learning is a powerful approach to construct a strong learner from multiple base learners. The most popular way to aggregate an ensemble of classifiers is majority voting, which assigns a sample to the class tha…

Brain Computer InterfaceEnsemble LearningGeneral Classification

Network Classifiers Based on Social Learning

2020-10-23 · Virginia Bordignon, Stefan Vlaski, Vincenzo Matta, Ali H. Sayed

This work proposes a new way of combining independently trained classifiers over space and time. Combination over space means that the outputs of spatially distributed classifiers are aggregated. Combination over time me…