Bayesian parameter estimation in exponential families
(30 minutes to learn)
Summary
Exponential families are convenient for Bayesian parameter estimation because the conjugate priors often have a convenient form, and there is a simple form for the posterior.
Context
This concept has the prerequisites:
Goals
- How do you derive the conjugate prior for an exponential family distribution?
- Show that the posterior can be computed in terms of the sufficient statistics.
- Work through a simple example, such as the beta-Bernoulli model.
Core resources (read/watch one of the following)
-Paid-
→ Machine Learning: a Probabilistic Perspective
A very comprehensive graudate-level machine learning textbook.
Location:
Section 9.2.5, "Bayes for the exponential family," pages 287-289
See also
-No Additional Notes-