Jethro's Braindump

Likelihood Principle

tags
Machine Learning

Given a generative model for data \(d\) given parameters \(\mathrm{\theta}\), \(P(d|\mathrm{\theta})\), and having observed a particular outcome \(d_1\), all inferences and predictions should depend only on the function \(P(d_1 | \mathrm{\theta})\).

Links to this note