In the previous blog post I covered the maximum likelihood method for parameter estimation in machine learning and statistical models. In this post we’ll go over another method for parameter estimation using Bayesian inference. I’ll also show how this method can be viewed as a generalisation of maximum likelihood and in what case the two methods are equivalent.
Some fundamental knowledge of probability theory is assumed e.g. marginal and conditional probability. These concepts are explained in my first post in this series. Additionally, it also helps to have some basic knowledge of a Gaussian distribution but it’s not necessary.