Corpus ID: 219531525

Balance-Subsampled Stable Prediction

  title={Balance-Subsampled Stable Prediction},
  author={Kun Kuang and H. Zhang and Fei Wu and Yueting Zhuang and Aijun Zhang},
  • Kun Kuang, H. Zhang, +2 authors Aijun Zhang
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • In machine learning, it is commonly assumed that training and test data share the same population distribution. However, this assumption is often violated in practice because the sample selection bias may induce the distribution shift from training data to test data. Such a model-agnostic distribution shift usually leads to prediction instability across unknown test data. In this paper, we propose a novel balance-subsampled stable prediction (BSSP) algorithm based on the theory of fractional… CONTINUE READING


    Publications referenced by this paper.
    Stable Prediction across Unknown Environments
    • 23
    • PDF
    Entropy Balancing for Causal Effects: A Multivariate Reweighting Method to Produce Balanced Samples in Observational Studies
    • 944
    • Highly Influential
    • PDF
    Approximate Residual Balancing: De-Biased Inference of Average Treatment Effects in High Dimensions.
    • 160
    • PDF
    The central role of the propensity score in observational studies for causal effects
    • 21,459
    • PDF
    Invariant Models for Causal Transfer Learning
    • 69
    • PDF
    A statistical perspective on algorithmic leveraging
    • 163
    • PDF
    Regression Shrinkage and Selection via the Lasso
    • 29,532
    • PDF
    Estimating Treatment Effect in the Wild via Differentiated Confounder Balancing
    • 22
    • PDF
    Treatment Effect Estimation with Data-Driven Variable Decomposition
    • 18
    • PDF