Pruning an ensemble of classifiers via reinforcement learning

Authors:
Ioannis Partalas;Grigorios Tsoumakas;Ioannis Vlahavas
Affiliations:
Department of Informatics, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece;Department of Informatics, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece;Department of Informatics, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece
Venue:
Neurocomputing
Year:
2009

Citing 25
Cited 5

The Strength of Weak Learnability

Machine Learning
Technical Note: \cal Q-Learning

Machine Learning
Original Contribution: Stacked generalization

Neural Networks
Bagging predictors

Machine Learning
An approach to the automatic design of multiple classifier systems

Pattern Recognition Letters - Special issue on machine learning and data mining in pattern recognition
A streaming ensemble algorithm (SEA) for large-scale classification

Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
Cost complexity-based pruning of ensemble classifiers

Knowledge and Information Systems
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
A Survey of Methods for Scaling Up Inductive Algorithms

Data Mining and Knowledge Discovery
Algorithm Selection using Reinforcement Learning

ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Pruning Adaptive Boosting

ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
Pruning and dynamic scheduling of cost-sensitive ensembles

Eighteenth national conference on Artificial intelligence
Mining concept-drifting data streams using ensemble classifiers

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Is Combining Classifiers with Stacking Better than Selecting the Best One?

Machine Learning
Ensemble selection from libraries of models

ICML '04 Proceedings of the twenty-first international conference on Machine learning
Pruning in ordered bagging ensembles

ICML '06 Proceedings of the 23rd international conference on Machine learning
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
Using boosting to prune bagging ensembles

Pattern Recognition Letters
Selective fusion of heterogeneous classifiers

Intelligent Data Analysis
Statistical Comparisons of Classifiers over Multiple Data Sets

The Journal of Machine Learning Research
Ensemble Pruning Via Semi-definite Programming

The Journal of Machine Learning Research
Reinforcement learning: a survey

Journal of Artificial Intelligence Research
Online adaptive policies for ensemble classifiers

Neurocomputing
Selective ensemble of decision trees

RSFDGrC'03 Proceedings of the 9th international conference on Rough sets, fuzzy sets, data mining, and granular computing
Ensemble pruning using reinforcement learning

SETN'06 Proceedings of the 4th Helenic conference on Advances in Artificial Intelligence

A fast ensemble pruning algorithm based on pattern mining process

Data Mining and Knowledge Discovery
Hybrid sampling on mutual information entropy-based clustering ensembles for optimizations

Neurocomputing
A probabilistic model of classifier competence for dynamic ensemble selection

Pattern Recognition
A measure of competence based on random classification for dynamic ensemble selection

Information Fusion
A competitive ensemble pruning approach based on cross-validation technique

Knowledge-Based Systems

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper studies the problem of pruning an ensemble of classifiers from a reinforcement learning perspective. It contributes a new pruning approach that uses the Q-learning algorithm in order to approximate an optimal policy of choosing whether to include or exclude each classifier from the ensemble. Extensive experimental comparisons of the proposed approach against state-of-the-art pruning and combination methods show very promising results. Additionally, we present an extension that allows the improvement of the solutions returned by the proposed approach over time, which is very useful in certain performance-critical domains.