Generalized Bayes Quantification Learning under Dataset Shift
From MaRDI portal
Abstract: Quantification learning is the task of prevalence estimation for a test population using predictions from a classifier trained on a different population. Quantification methods assume that the sensitivities and specificities of the classifier are either perfect or transportable from the training to the test population. These assumptions are inappropriate in the presence of dataset shift, when the misclassification rates in the training population are not representative of those for the test population. Quantification under dataset shift has been addressed only for single-class (categorical) predictions and assuming perfect knowledge of the true labels on a small subset of the test population. We propose generalized Bayes quantification learning (GBQL) that uses the entire compositional predictions from probabilistic classifiers and allows for uncertainty in true class labels for the limited labeled test data. Instead of positing a full model, we use a model-free Bayesian estimating equation approach to compositional data based only on a first-moment assumption. The idea will be useful in Bayesian compositional data analysis in general as it is robust to different generating mechanisms for compositional data and includes categorical outputs as a special case. We show how our method yields existing quantification approaches as special cases. Extension to an ensemble GBQL that uses predictions from multiple classifiers yielding inference robust to inclusion of a poor classifier is discussed. We outline a fast and efficient Gibbs sampler using a rounding and coarsening approximation to the loss functions. We also establish posterior consistency, asymptotic normality and valid coverage of interval estimates from GBQL, as well as finite sample posterior concentration rate. Empirical performance of GBQL is demonstrated through simulations and analysis of real data with evident dataset shift.
Recommendations
- Bayesian parameter learning with an application
- scientific article; zbMATH DE number 7219345
- Bayesian learning with Wasserstein barycenters
- Discriminative learning under covariate shift
- Generalised data augmentation and posterior inferences
- Posterior concentration and fast convergence rates for generalized Bayesian learning
- Empirical Bayesian Learning
- Input-dependent estimation of generalization error under covariate shift
Cites work
- scientific article; zbMATH DE number 6667653 (Why is no real title available?)
- scientific article; zbMATH DE number 47310 (Why is no real title available?)
- scientific article; zbMATH DE number 7272386 (Why is no real title available?)
- A General Framework for Updating Belief Distributions
- An MCMC approach to classical estimation.
- Assigning a value to a power likelihood in a general Bayesian model
- Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables
- Bayesian fractional posteriors
- Calibrating general posterior credible regions
- From -entropy to KL-entropy: analysis of minimum information complexity density estima\-tion
- Generalized Dirichlet distribution in Bayesian analysis
- Gibbs posterior for variable selection in high-dimensional classification and data mining
- Identifiability of Finite Mixtures
- Inconsistency of Bayesian inference for misspecified linear models, and a proposal for repairing it
- Longitudinal data analysis using generalized linear models
- Modeling compositional data using Dirichlet regression models
- Multivariate fractional regression estimation of econometric share models
- On Bayesian consistency
- On the Identifiability of Finite Mixtures
- Robust Bayesian inference via coarsening
- Some PAC-Bayesian theorems
- The Bernstein-von Mises theorem under misspecification
- The Continual Reassessment Method for Multiple Toxicity Grades: A Bayesian Quasi‐Likelihood Approach
- The safe Bayesian. Learning the learning rate via the mixability gap
- Verbal autopsy methods with multiple causes of death
Cited in
(7)- Modeling structure and country-specific heterogeneity in misclassification matrices of verbal autopsy-based cause of death classifiers
- Prior shift using the ratio estimator
- Quantification-oriented learning based on reliable classifiers
- Aggregative quantification for regression
- A transformation‐free linear regression for compositional outcomes and predictors
- Binary quantification and dataset shift: an experimental investigation
- Quantification under prior probability shift: the ratio estimator and its extensions
This page was built for publication: Generalized Bayes Quantification Learning under Dataset Shift
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6110729)
