Bias Correction for Regularized Regression and its Application in Learning with Streaming Data
This work addresses bias reduction for regularized regression, offering incremental improvements for applications in streaming data learning.
The paper tackles bias in ridge regression and regularization kernel networks, proposing new algorithms that achieve comparable performance on single datasets and greater efficiency in incremental learning with streaming data, as verified by theory and simulations.
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.