This course and its lessons is the reason why I created CauseCade. I wish to teach the concept of bayesian networks to a wide audience using an interactive experience. I believe that by teaching material in this way, students will be able to learn at their own pace and get a better feel for the material.
I will continuously be updating both the written material and the functionality of CauseCade, focusing primarily on enhancing the experience for students. With this in mind, I hope that many people provide with feedback on sections that they think could use improvement or be expandend. For those who already have previous experience with Bayesian Networks and wish to contribute some of their knowledge or examples, this would be of great help. Contact info can be found on the blog or in the help menu.
I think it would be wise to explain the general concept and merits of bayesian networks before diving into the underlying theory and operating equations. Depending on what discipline of science you may be interested in, you may have heard of bayesian networks. If you haven't', you will probably have heard of Bayes' rule. (If not, I recommend you visit the probability course first - or wait until you taken a course covering the basic concepts of probability first).
As the name suggests, bayesian network make use of Bayes' rule. <(Perhaps somewhat unexpectedly, it is not the use of Bayes rule that is what makes bayesian networks so useful.)> This does not mean, however, that one must abandon all frequentist notions and solely dedicate oneself bayesian statistics. If the term frequentist and bayesianism sound foreign to you, rest assured, it is not central to the following sections. They are two different views on statistics, and are rather interesting to read up on.
Fundamentally, bayesian network chose to represent uncertainty in the framework of probability. There are other ways to model uncertainty, but bayesian networks chose to do this with the system that we are familiar with. <(missing lines)>
In the previous course, we derived Bayes' rules for general random variables. In this course, we will be putting this rule to work in a network form, building up from simple beginnings. Along the way, we shall put particular emphasis on the implications and consequences of the formulas that we will use.