Mitigating Outlier Effect in Online Regression: An Efficient Usage of Error Correntropy Criterion

Bahrami, Sajjad; Tuncel, Ertem

doi:10.1109/ijcnn48605.2020.9207141

Cited by 4 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It indeed determines the magnitude of the weights assigned to each error sample and it is a function of error. Optimizing this bandwidth has been widely addressed in previous work, for instance by minimizing Kullback-Leibler divergence between the true and estimated error distribution, using shape of error distribution measured by its kurtosis, using instantaneous error in each iteration, changing the Gaussian kernel, using hybrid methods and so forth [34,35,36,37,38,39,40,41].…”

Section: A Overview Of MCC and Meementioning

confidence: 99%

Trimmed Minimum Error Entropy for Robust Online Regression

Bahrami¹,

Tuncel²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, online linear regression in environments corrupted by non-Gaussian noise (especially heavytailed noise) is addressed. In such environments, the error between the system output and the label also does not follow a Gaussian distribution and there might exist abnormally large error samples (or outliers) which mislead the learning process. The main challenge is how to keep the supervised learning problem least affected by these unwanted and misleading outliers.In recent years, an information theoretic algorithm based on Renyi's entropy, called minimum error entropy (MEE), has been employed to take on this issue. However, this minimization might not result in a desired estimator inasmuch as entropy is shift-invariant, i.e., by minimizing the error entropy, error samples may not be necessarily concentrated around zero. In this paper, a quantization technique is proposed by which not only aforementioned need of setting errors around the origin in MEE is addressed, but also major outliers are rejected from MEE-based learning and MEE performance is improved from convergence rate, steady state misalignment, and testing error points of view.

show abstract