Models of Learning

Developed by
Jill O'Reilly
Hanneke den Ouden
-August 2015

Marginal distributions for α and β

Most of the time we will be interested in the value of each parameter, not in this full 2D 'likelihood landscape'.

In order to achieve this, we can for each parameter, or dimension, 'collapse' or 'marginalise' this plot over the other dimensions.

What do you think the marginal distributions for α and β look like?