Discriminative Density-ratio Estimation

TL;DR

This paper introduces a novel discriminative density-ratio estimation method that improves covariate shift adaptation by preserving class separation, leading to better accuracy and robustness in classification tasks.

Contribution

The paper proposes a class-wise iterative density-ratio estimation method with a soft matching technique and mutual information-based stopping criterion.

Findings

Outperforms existing methods in accuracy on synthetic and benchmark datasets.

Demonstrates robustness to covariate shift in classification tasks.

Effectively preserves class separation during reweighting.

Abstract

The covariate shift is a challenging problem in supervised learning that results from the discrepancy between the training and test distributions. An effective approach which recently drew a considerable attention in the research community is to reweight the training samples to minimize that discrepancy. In specific, many methods are based on developing Density-ratio (DR) estimation techniques that apply to both regression and classification problems. Although these methods work well for regression problems, their performance on classification problems is not satisfactory. This is due to a key observation that these methods focus on matching the sample marginal distributions without paying attention to preserving the separation between classes in the reweighted space. In this paper, we propose a novel method for Discriminative Density-ratio (DDR) estimation that addresses the aforementioned problem and aims at estimating the density-ratio of joint distributions in a class-wise manner. The proposed algorithm is an iterative procedure that alternates between estimating the class information for the test data and estimating new density ratio for each class. To incorporate the estimated class information of the test data, a soft matching technique is proposed. In addition, we employ an effective criterion which adopts mutual information as an indicator to stop the iterative procedure while resulting in a decision boundary that lies in a sparse region. Experiments on synthetic and benchmark datasets demonstrate the superiority of the proposed method in terms of both accuracy and robustness.