ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2305.08637
22
7

Double-Weighting for Covariate Shift Adaptation

15 May 2023
José I. Segovia-Martín
Santiago Mazuelas
Anqi Liu
ArXivPDFHTML
Abstract

Supervised learning is often affected by a covariate shift in which the marginal distributions of instances (covariates xxx) of training and testing samples ptr(x)\mathrm{p}_\text{tr}(x)ptr​(x) and pte(x)\mathrm{p}_\text{te}(x)pte​(x) are different but the label conditionals coincide. Existing approaches address such covariate shift by either using the ratio pte(x)/ptr(x)\mathrm{p}_\text{te}(x)/\mathrm{p}_\text{tr}(x)pte​(x)/ptr​(x) to weight training samples (reweighted methods) or using the ratio ptr(x)/pte(x)\mathrm{p}_\text{tr}(x)/\mathrm{p}_\text{te}(x)ptr​(x)/pte​(x) to weight testing samples (robust methods). However, the performance of such approaches can be poor under support mismatch or when the above ratios take large values. We propose a minimax risk classification (MRC) approach for covariate shift adaptation that avoids such limitations by weighting both training and testing samples. In addition, we develop effective techniques that obtain both sets of weights and generalize the conventional kernel mean matching method. We provide novel generalization bounds for our method that show a significant increase in the effective sample size compared with reweighted methods. The proposed method also achieves enhanced classification performance in both synthetic and empirical experiments.

View on arXiv
Comments on this paper