If you haven’t already, please start to think about what to do for your final project and with whom. (Individual projects are fine, but group projects are also encouraged.) Below are two things you can do; they are not mutually exclusive:

• Write and explain a program to model a stochastic process, probably a cognitive process such as perception or action. The model should preferably be causal: running the model should simulate the process.

• Experiment with a computation that consumes or produces a model, such as probabilistic inference. Explore the consequence of the model and the behavior of the computation, perhaps by visualizing them in action. Discuss how well they match your intuition and physical experiments.

On the last day of class, Wednesday April 25, will be a presentation for each project. Please tell me ahead of time how much time you need for your presentation. You also need to write a paper to describe your project. How about making the paper due on Wednesday May 2? Both the project and the paper should target your classmates as their audience and be as short as possible but no shorter.

By the way, you can find more details on Bayesian learning in Heckerman’s tutorial.