Invariance assumptions for class distribution estimation
Author:
Preprint
Abstract
We study the problem of class distribution estimation under dataset shift. On the training dataset, both features and class labels are observed while on the test dataset only the features can be observed. The task then is the estimation of the distribution of the class labels, i.e. the estimation of the class prior probabilities, in the test dataset. Assumptions of invariance between the training joint distribution of features and labels and the test distribution can considerably facilitate this task. We discuss the assumptions of covariate shift, factorizable joint shift, and sparse joint shift and their implications for class distribution estimation.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Learning to Quantify2023
- Marginal singularity and the benefits of labels in covariate-shiftThe Annals of Statistics, 2021
- Calibrating sufficientlyStatistics, 2021
- Temporal density extrapolation using a dynamic basis approachData Mining and Knowledge Discovery, 2019
- Adapting a classification rule to local and global shift when only unlabelled data are availableEuropean Journal of Operational Research, 2015
- Semi-supervised learning of class balance under class-prior change by distribution matchingNeural Networks, 2014
- Density Ratio Estimation in Machine Learning2012
- A Response to Webb and Ting?s On the Application of ROC Analysis to Predict Classification Performance Under Varying Class DistributionsMachine Learning, 2005
- Adjusting the Outputs of a Classifier to New aPrioriProbabilities: A Simple ProcedureNeural Computation, 2002
- Improving predictive inference under covariate shift by weighting the log-likelihood functionJournal of Statistical Planning and Inference, 2000