Adjusting the Outputs of a Classifier to New a Priori Probabilities: A Simple Procedure

By Marco Saerens, Patrice Latinne, Christine Decaestecker. 2002

463 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering, Statistics and Probability

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

In Statistics and Probability

Citations in the corpus

Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift (2018, background): “Owing to the impossibility of correcting for shift absent assumptions [3], these papers often assume either covariate shift q(x, y) = q(x)p(y|x) [15, 45, 49] or label shift q(x, y) = q(y)p(x|y) [7, 29, 38, 48, 56].”
Detecting and Correcting for Label Shift with Black Box Predictors (2018, background): “Post-submission, we learned of antecedents for our estimator in epidemiology (Buck et al., 1966) and revisited by Forman (2008); Saerens et al. (2002).”
Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment (2019, background): “Some consider the reverse setting label shift (Saerens et al., 2002; Zhang et al., 2013; Lipton et al., 2018), where although the label distribution shifts ( p S ( y ) (cid:54) = p T ( y )), the class-conditional input distribution is invariant ( p S ( x | y ) = p T ( x | y )).”
Domain Adaptive Transfer Learning with Specialist Models (2018, methodology): “related3 to prior probability shift [28, 31] also known as target shift [39].”
WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020, neutral): “…e.g., learning to map data points from source to target domains (Hoffman et al., 2018), or estimating the test label distribution from unlabeled test data (Saerens et al., 2002; Zhang et al., 2013; Lipton et al., 2018; Azizzadenesheli et al., 2019; Alexandari et al., 2020; Garg et al., 2020).”
Calibrating Probability with Undersampling for Unbalanced Classification (2015, methodology): “[21] show that, given different priors between the training and testing sets, the posterior probability can be corrected with the following equation:”
An introduction to domain adaptation and transfer learning (2018, background): “The knowledge that the conditionals are equivalent can be exploited by canceling them out of the ratio of joint probability distributions [165]:”
Classifier adaptation at prediction time (2015, methodology): “If π were known, we could immediately derive an optimal adaptation rule in closed form, see [23] for a detailed derivation.”
The importance of intrinsic disorder for protein phosphorylation. (2004, unknown): “Predictors trained using a specific class distribution are known not to be optimal when applied to unlabeled data with different class priors ( 37 ).”
Word Sense Disambiguation with Distribution Estimation (2005, methodology): “To estimate the sense distribution, or a priori probabilities of the different senses in a new data set, we made use of a confusion matrix algorithm [Vucetic and Obradovic, 2001] and an EM based algorithm [Saerens et al., 2002], which we describe in this section.”
The Open World of Micro-Videos (2016, background): “For example, if we assume that the popularity of tags changes over time but their appearance models do not, one can model eﬃciently model temporal dynamics with a statistical prior shift [42].”
Quantifying counts and costs via classification (2008, methodology): “…the less common work that acknowledges that the training and test distributions may differ, the estimate of the class distribution is used only in order to improve the individual classifications, e.g. via ROC analysis (Provost and Fawcett 2001; Fawcett 2003; Mei and Zhai 2005; Saerens et al. 2002).”
Machine Learning for 5G/B5G Mobile and Wireless Communications: Potential, Limitations, and Future Directions (2019, background): “This adaptation may be done either by the end user [43], [44], or automatically [45], [46].”
A Comprehensive Survey on Test-Time Adaptation Under Distribution Shifts (2023, unknown): “Other distribution shifts like prior shift [32] are further discussed in Sec.”
ReAct: Out-of-distribution Detection With Rectified Activations (2021, background): “Meanwhile, some works considered label distribution shift [1, 2, 32, 42, 47, 53], where the label space is common between ID and OOD but the marginal label distribution changes, as well as covariate shift in the input space [14, 39], where inputs can be corruption-shifted or domainshifted [17, 48].”