Neural Conditional Probability for Inference

• Main Authors: Kostic, Vladimir R, Lounici, Karim, Pacreau, Gregoire, Novelli, Pietro, Turri, Giacomo, Pontil, Massimiliano
• Format: Journal Article
• Language: English
• Published: 01.07.2024
• Subjects: Computer Science - Learning; Mathematics - Statistics Theory; Statistics - Machine Learning; Statistics - Methodology; Statistics - Theory

We introduce NCP (Neural Conditional Probability), a novel operator-theoretic approach for learning conditional distributions with a particular focus on inference tasks. NCP can be used to build conditional confidence regions and extract important statistics like conditional quantiles, mean, and cov...