Robust Learning under Uncertain Test Distributions: Relating Covariate Shift to Model Misspecification

@inproceedings{Wen2014RobustLU,
  title={Robust Learning under Uncertain Test Distributions: Relating Covariate Shift to Model Misspecification},
  author={Junfeng Wen and Chun-Nam Yu and Russell Greiner},
  booktitle={ICML},
  year={2014}
}
Many learning situations involve learning the conditional distribution ppy|xq when the training instances are drawn from the training distribution ptrpxq, even though it will later be used to predict for instances drawn from a different test distribution ptepxq. Most current approaches focus on learning how to reweigh the training examples, to make them resemble the test distribution. However, reweighing does not always help, because (we show that) the test error also depends on the correctness… CONTINUE READING
Highly Cited
This paper has 37 citations. REVIEW CITATIONS

Citations

Publications citing this paper.

References

Publications referenced by this paper.
Showing 1-10 of 27 references

The performance of both linear model and cubic model are investigated for this covariate shift scenario

Shimodaira
Figure 7 reports the test losses (mean and one standard deviation as errorbar) of different reweighing algorithms over 10 trials. Obviously, reweighing is relatively • 2000
View 8 Excerpts
Highly Influenced

Convex analysis

R. T. Rockafellar
1996
View 1 Excerpt
Highly Influenced

Consequences and detection of misspecified nonlinear regression models

H. White
Journal of the American Statistical Association, • 1981
View 3 Excerpts
Highly Influenced

Dataset shift in machine learning

J. Quionero-Candela, M. Sugiyama, A. Schwaighofer, N. D. Lawrence
2009
View 2 Excerpts

Discriminative Learning Under Covariate Shift

Journal of Machine Learning Research • 2009
View 1 Excerpt