Machine Learning in Non-Stationary Environments - Introduction to Covariate Shift Adaptation

By Masashi Sugiyama, Motoaki Kawanabe. 2012

132 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Control and Systems Engineering

Citations in the corpus

Theoretical evidence for adversarial robustness through randomization: the case of the Exponential family (2019, background): “However, simply doing so ruins the accuracy of the algorithm (as it becomes an instance of distribution shift [44]).”
On causal and anticausal learning (2012, background): “This scenario is referred to as covariate shift (Sugiyama & Kawanabe, 2012).”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, background): “cost of a model at training time, but also at test time can be a limiting factor (e.g., for data streams or in real-time monitoring). Is the data-generating process assumed to be non-stationary [239]–[241] and are there distributional shifts expected at test time? For (truly) high-dimensional data, ”
Domain Aggregation Networks for Multi-Source Domain Adaptation (2019, background): “been explored extensively for the last decade using different assumptions (Pan and Yang, 2009; Zhang et al., 2015). For instance, the covariate shift scenario (Shimodaira, 2000; Gretton et al., 2009; Sugiyama and Kawanabe, 2012; Wen et al., 2014) assumes S6= Tbut f S= f T, while concept drift (Jiang”
Density-ratio matching under the Bregman divergence: a unified framework of density-ratio estimation (2012, background): “2009, 2012) such as discriminant analysis (Silverman 1978), non-stationarity adaptation (Shimodaira 2000; Sugiyama and Müller 2005; Sugiyama et al. 2007; Quiñonero-Candela et al. 2009; Sugiyama and Kawanabe 2011), multitask learning (Bickel et al.”
Importance-weighted least-squares probabilistic classifier for covariate shift adaptation with application to human activity recognition (2012, background): “Popular lines of research include re-weighting samples according to the importance [23, 12, 30, 13, 6, 27] and learning feature representation that is common to training and test data [4, 8, 3, 9, 18].”
Survey on supervised machine learning techniques for automatic text classification (2019, background): “According to Sugiyama and Kawanabe (2012) enhanced k-NN classifier for text categorization, which builds the categorization pattern utilizing mixing, limited to one badge gathering process and k-NN categorization.”
Robust Learning under Uncertain Test Distributions: Relating Covariate Shift to Model Misspecification (2014, methodology): “To correct the shifted distribution, major efforts have been dedicated to importance reweighing (Quionero-Candela et al., 2009; Sugiyama & Kawanabe, 2012).”