Optimal Linear Combination of Classifiers

The question of whether to use one classifier or a combination of classifiers is a central topic in Machine Learning. We propose here a method for finding an optimal linear combination of classifiers derived from a bias-variance framework for the classification task.

cs.LG cs.AI Neural Networks and Applications Face and Expression Recognition Fuzzy Logic and Control Systems

Source

Paper

Similar papers

Analytical study of the optimal combination of binary classifiers based on classifiers-induced partitioning of the training setJ. Brossier, O. Lafitte2026 · 0 citations In Library

Learning Mixtures of Linear ClassifiersYuekai Sun, Stratis Ioannidis, A. Montanari2013 · International Conference on Machine Learning · 23 citations In Library

Combining Classifiers for Improved Accuracies -Voting and Linearly Weighted AlgorithmsAlbert Wong, Athena Nguyen, Eugene Li et al.2026 In Library

Linear Classifier Combination via Multiple Potential FunctionsPawel Trajdos, R. Burduk2020 · Pattern Recognition · 8 citations In Library

Integration and Selection of Linear SVM Classifiers in Geometric SpaceJędrzej Biedrzycki, Robert Burduk2020 · DOAJ (DOAJ: Directory of Open Access Journals) In Library

Optimal Feature Selection for Support Vector Machine ClassifiersOliver Strub2020 · IEEE International Conference on Industrial Engineering and Engineering Management · 1 citation In Library

Application of an Ensemble of Linear Separators for a Universal Decision in Multiclass ClassificationPavel Chernavin, Nikolai Chernavin, Fedor Chernavin2025 In Library

Combining One-Class Classifiers via Meta-LearningE. Menahem, L. Rokach, Y. Elovici2011 · International Conference on Information and Knowledge Management · 35 citations In Library

Demystifying the Optimal Fair Classifier in Multi-Class ClassificationLi Zhang, Yuyuan Li, Xiaohua Feng et al.2026 · 0 citations In Library

Optimally Combining Classifiers for Semi-Supervised LearningZhiguo Wang, Liusha Yang, Feng Yin et al.2020 · arXiv (Cornell University) · 2 citations In Library

Patents near this paper

DECISION MAKING IN CLASSIFICATION PROBLEMSUS 6,931,351 · INTERNATIONAL BUSINESS MACHINES CORPORATION · 2005 WEIGHTED ENSEMBLE BOOSTING METHOD FOR CLASSIFIER COMBINATION AND FEATURE SELECTIONUS 7,529,403 · MITSUBISHI ELECTRIC RESEARCH LABORATORIES, INC. · 2009 CONFIDENCE WEIGHTED CLASSIFIER COMBINATION FOR MULTI-MODAL IDENTIFICATIONUS 7,742,641 · HONDA MOTOR CO., LTD · 2010 METHOD FOR COMBINING BOOSTED CLASSIFIERS FOR EFFICIENT MULTI-CLASS OBJECT DETECTIONUS 7,769,228 · SIEMENS CORPORATE RESEARCH, INC. · 2010 CLASSIFICATION GENERATION METHOD USING COMBINATION OF MINI-CLASSIFIERS WITH REGULARIZATION AND USES THEREOFUS 9,477,906 · BIODESIX, INC. · 2016

Source of record

arXiv