Bias Correction for Regularized Regression and its Application in Learning with Streaming Data

TL;DR

This paper introduces bias correction methods for ridge regression and regularization kernel networks, improving efficiency in streaming data scenarios while maintaining comparable performance to traditional methods.

Contribution

It presents novel bias correction algorithms for regularized regression models, enhancing incremental learning efficiency in streaming data contexts.

Findings

Bias correction reduces estimation bias in ridge regression and kernel networks.

The new algorithms are more efficient in streaming data applications.

Theoretical analysis and simulations confirm the effectiveness of the proposed methods.

Abstract

We propose an approach to reduce the bias of ridge regression and regularization kernel network. When applied to a single data set the new algorithms have comparable learning performance with the original ones. When applied to incremental learning with block wise streaming data the new algorithms are more efficient due to bias reduction. Both theoretical characterizations and simulation studies are used to verify the effectiveness of these new algorithms.