Input-dependent estimation of generalization error under covariate shift

By Masashi Sugiyama, Muller Klaus-Robert. 2005

158 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Mechanics, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Mechanics

In Control and Systems Engineering

Citations in the corpus

Data Augmentation Generative Adversarial Networks (2017, background): “The term dataset shift (Storkey, 2009) generalises the concept of covariate shift (Shimodaira, 2000; Storkey & Sugiyama, 2007; Sugiyama & Müller, 2005) to multiple cases of changes between domains.”
Covariate Shift Adaptation by Importance Weighted Cross Validation (2007, methodology): “Model selection under the covariate shift has been studied so far only by few researchers (e.g., Shimodaira, 2000; Sugiyama and Müller, 2005)—existing methods have a number of limitations, e.”
A Brief Review of Domain Adaptation (2020, methodology): “To solve this problem, [44], proposed a new model that uses a linear model w(x) = φ(x)α, where φ(x) is a basis function and α is a set of parameters to be learned.”
Density-ratio matching under the Bregman divergence: a unified framework of density-ratio estimation (2012, background): “, 2009, 2012) such as discriminant analysis (Silverman, 1978), nonstationarity adaptation (Shimodaira, 2000; Sugiyama and Müller, 2005; Sugiyama et al., 2007; Quiñonero-Candela et al., 2009; Sugiyama and Kawanabe, 2011), multi-task learning (Bickel et al.”
Towards adaptive classification for BCI (2006, background): “This could in principle make classification a difficult problem, perhaps necessitating special statistical modeling that takes into account covariate shifts [29] or even more sophisticated techniques such as transductive inference [30].”
When Training and Test Sets are Different: Characterising Learning Transfer (2013, methodology): “This was extended in [26], with the inclusion of a generalisation error estimation method for this process of adapting for covariate shift.”
Localized Conformal Prediction: A Generalized Inference Framework for Conformal Prediction (2021, background): “(2019a), the authors consider average coverage for the test sample distribution, given that we have enough test samples to estimate the covariate shift (Shimodaira, 2000; Sugiyama and Müller, 2005; Sugiyama et al., 2007; Quionero-Candela et al., 2009).”
Domain Adaptation Problems: A DASVM Classification Technique and a Circular Validation Strategy (2010, background): “On the other hand, the particular case where the true and estimated distributions are assumed to differ only via ^ P ðxÞ 6 ¼ P ðxÞ, but ^ P ðy j xÞ % P ðy j xÞ is denoted by covariate shift [7], [8].”
Direct importance estimation for covariate shift adaptation (2008, background): “However, the ordinary CV score is heavily biased due to covariate shift and we also need to importance-weight the CV score (or any other model selection criteria) for unbiasedness (Shimodaira, 2000; Sugiyama and Müller, 2005; Sugiyama et al., 2007).”