Dynamic Bayesian combination of multiple imperfect classifiers

DOI10.1007/978-3-642-36406-8_1MaRDI QIDQ2822296zbMATHOpenAlexFDO

Authors Edwin Simpson, Stephen J. Roberts

Publication date 30 September 2016

Published in Decision Making and Imperfection (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1206.1831

Mathematics Subject Classification ID

Bayesian inference (62F15) Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10)

Abstract: Classifier combination methods need to make best use of the outputs of multiple, imperfect classifiers to enable higher accuracy classifications. In many situations, such as when human decisions need to be combined, the base decisions can vary enormously in reliability. A Bayesian approach to such uncertain combination allows us to infer the differences in performance between individuals and to incorporate any available prior knowledge about their abilities when training data is sparse. In this paper we explore Bayesian classifier combination, using the computationally efficient framework of variational Bayesian inference. We apply the approach to real data from a large citizen science project, Galaxy Zoo Supernovae, and show that our method far outperforms other established approaches to imperfect decision combination. We go on to analyse the putative community structure of the decision makers, based on their inferred decision making strategies, and show that natural groupings are formed. Finally we present a dynamic Bayesian classifier combination approach and investigate the changes in base classifier performance over time.

Recommendations

• A Bayesian framework for the combination of classifier outputs
• The combination of multiple classifiers using an evidential reasoning approach
• Robust naive Bayes combination of multiple classifications
• scientific article; zbMATH DE number 1860779
• Combining Classifiers via Discretization

Cites work

• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 3097528 (Why is no real title available?)
• Adaptive Rejection Sampling for Gibbs Sampling
• Bayesian Inference on Multiscale Models for Poisson Intensity Estimation: Applications to Photon-Limited Image Denoising
• Bayesian forecasting and dynamic models.
• Community structure in social and biological networks
• Dynamic Generalized Linear Models and Bayesian Forecasting
• Fixed interval smoothing with discrete measurements
• On Information and Sufficiency
• Pattern recognition and machine learning.
• Statistical field theory. With a foreword by David Pines
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
• The weighted majority algorithm
• Variational Mixture of Bayesian Independent Component Analyzers

Cited in

(7)

• A joint Gaussian process model for active visual recognition with expertise estimation in crowdsourcing
• scientific article; zbMATH DE number 7049770 (Why is no real title available?)
• Quality-aware online task assignment mechanisms using latent topic model
• Learning from crowds with sparse and imbalanced annotations
• Bayesian methods for intelligent task assignment in crowdsourcing systems
• Approximate Bayesian recursive estimation
• A Bayesian framework for the combination of classifier outputs

This page was built for publication: Dynamic Bayesian combination of multiple imperfect classifiers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2822296)