Concentration and Calibration in Predictive Bayesian Inference

Name: Concentration and Calibration in Predictive Bayesian Inference
Uploaded: 2026-10-07T21:39:02.580033+00:00
Duration: 317 s
Description: Predictive Bayesian Inference (PBI) attempts to infer population functionals using only a predictive engine that generates future data. While PBI posteriors often concentrate onto a well-defined quantity, the resulting uncertainty quantification can be inaccurate if the predictive engine fails to accurately capture all relevant features of the underlying data distribution.

Predictive Bayesian Inference (PBI) attempts to infer population functionals using only a predictive engine that generates future data. While PBI posteriors often concentrate onto a well-defined quantity, the resulting uncertainty quantification can be inaccurate if the predictive engine fails to accurately capture all relevant features of the underlying data distribution.

Open the full episode