Uporedite metode

	Konformalno predviđanje	Kalibracija modela
Oblast	Mašinsko učenje	Mašinsko učenje
Porodica	Machine learning	Machine learning
Godina nastanka	2005	2017
Tvorac	Vovk, Gammerman & Shafer	Platt; Guo et al.
Tip	Distribution-free uncertainty quantification framework	Post-hoc probability correction technique
Temeljni izvor	Vovk, V., Gammerman, A., & Shafer, G. (2005). Algorithmic Learning in a Random World. Springer. ISBN: 978-0-387-00152-4	Guo, C., Pleiss, G., Sun, Y., & Weinberger, K. Q. (2017). On calibration of modern neural networks. International Conference on Machine Learning, 1321–1330.
Drugi nazivi	Conformal Inference, Conformal Risk Control, Inductive Conformal Prediction, Uyumsal Tahmin	Classifier Calibration, Probability Calibration, Score Calibration, Model Kalibrasyonu
Srodne	2	3
Sažetak	Conformal Prediction is a distribution-free framework for constructing statistically valid prediction sets (for classification) or prediction intervals (for regression) around the output of any pre-trained machine learning model. Introduced by Vovk, Gammerman, and Shafer in their 2005 monograph, it provides a finite-sample marginal coverage guarantee — the true label falls inside the prediction set with at least 1-alpha probability — without requiring parametric assumptions about the data distribution.	Model calibration is a post-hoc technique that adjusts the probability outputs of a trained classifier so that predicted confidence scores match empirical outcome frequencies. A classifier is said to be perfectly calibrated if, among all predictions made with confidence p, exactly a fraction p of them are correct. Systematic miscalibration of modern deep neural networks was rigorously documented by Guo et al. (2017), who showed that networks trained with standard cross-entropy loss tend to be overconfident, and proposed temperature scaling as a simple, effective remedy.