Glossario IA
Il dizionario completo dell'Intelligenza Artificiale
162
categorie
2.032
sottocategorie
23.060
termini
termini
Probability Binning
A discretization process that groups probabilistic predictions into contiguous intervals to facilitate the analysis and visualization of calibration at different confidence levels.
termini
Decision Calibration
A specific type of calibration that ensures binary decisions based on a probability threshold maintain an error rate consistent with the predicted probabilities.
termini
Bayesian Binning
A probabilistic calibration method that uses Bayesian principles to estimate the posterior distributions of error rates in each bin, incorporating uncertainty.
termini
Prior Calibration
The process of adjusting the model's a priori probabilities to match the actual class distributions in the target population.
🔍