Quantile regression (QR) Machine Learning; Probabilistic / Bayesian AI
noun phrase
Definition: A regression method that estimates conditional quantiles of a response variable, rather than only its conditional mean, and is therefore widely used for probabilistic prediction and uncertainty-aware modeling [Papacharalampous, Langousis 2022].
Example in context: “Quantile regression is a type of regression method widely used in statistics, econometrics, reinforcement learning, and large language models.” [Lin et al. 2024]
Synonym: QR
Related terms: quantile regression forest; probabilistic regression; quantile loss; conditional quantile estimation