Having seen a probabilistic view for classification, and some of the computational difficulties in higher order approximations in both regression and classification, we now turn toward a simple, yet general and highly powerful framework for regression and clas…

In recent years many different industries have been drawn toward the techniques of machine learning and artificial intelligence to solve various real world problems that are not easily attacked by simple algorithms or analytic methods. Examples of such problem…

Classification Redux Now that we have had a look at gradient descent models and perdorming regression using a cost function, I would like to return to the concept of classification. Let us say that, instead of smooth, contiguous dataset, we have something mor…

I'd like to depart, for a bit, from the notion of classification so that we can introduce another important part of machine learning--regression. It is a common problem that one collects a set of data that is, perhaps, a bit noisy, and still wishes to find som…

Some basic probability theory To begin a discussion of stochastic modeling, it is important to note some basic properties of probability and define some useful notation. We will then take a basic amount of this information, and apply it very trivially to get a…