Restricted Minimum Error Entropy Criterion for Robust Classification

TL;DR

This paper introduces a restricted minimum error entropy (RMEE) criterion that enhances robustness in classification tasks, especially under noisy conditions, by guiding the error distribution towards a desired form.

Contribution

The paper proposes the RMEE method, which modifies the original MEE to improve robustness in noisy classification by constraining the error distribution.

Findings

RMEE outperforms traditional MEE in noisy environments.

Experimental results with logistic regression and ELM validate robustness.

The method converges reliably under half-quadratic optimization.

Abstract

The minimum error entropy (MEE) criterion has been verified as a powerful approach for non-Gaussian signal processing and robust machine learning. However, the implementation of MEE on robust classification is rather a vacancy in the literature. The original MEE only focuses on minimizing the Renyi's quadratic entropy of the error probability distribution function (PDF), which could cause failure in noisy classification tasks. To this end, we analyze the optimal error distribution in the presence of outliers for those classifiers with continuous errors, and introduce a simple codebook to restrict MEE so that it drives the error PDF towards the desired case. Half-quadratic based optimization and convergence analysis of the new learning criterion, called restricted MEE (RMEE), are provided. Experimental results with logistic regression and extreme learning machine are presented to verify the desirable robustness of RMEE.