Introductory paper on Hierarchical Bayes Multitask Learning?

I'm interested in doing a class project on multitask learning, but the dataset I want to use isn't large enough to do an approach using a neural net with multiple outputs. Therefore I'd like to do a Hierarchical Bayes approach where I impose a shared prior on the weights for all of the tasks. I was wondering if anyone has a recommendation for a simple, clear paper that discusses how to do this for regression, for example. The papers I've been able to find seem overly complex for what I'd like to do.