Robust Correction of Sampling Bias Using Cumulative Distribution Functions
This addresses covariate shift correction for machine learning practitioners dealing with biased datasets, though it appears incremental as it builds on a recent idea.
The paper tackles the problem of covariate shift between training and target distributions by introducing a new method based on empirical cumulative distribution functions, which experimentally demonstrates robustness without parameter tuning while maintaining similar classification performance to state-of-the-art techniques.
Varying domains and biased datasets can lead to differences between the training and the target distributions, known as covariate shift. Current approaches for alleviating this often rely on estimating the ratio of training and target probability density functions. These techniques require parameter tuning and can be unstable across different datasets. We present a new method for handling covariate shift using the empirical cumulative distribution function estimates of the target distribution by a rigorous generalization of a recent idea proposed by Vapnik and Izmailov. Further, we show experimentally that our method is more robust in its predictions, is not reliant on parameter tuning and shows similar classification performance compared to the current state-of-the-art techniques on synthetic and real datasets.