Variational Inference

less than 1 minute read

Published:

Topics: Variational Inference

Variational inference is an important topic that is widely used in machine learning. For example, it’s the basis for variational autoencoders. Also Bayesian learning often makes use variational of inference. To understand what variational inference is, how it works and why it’s useful I went through each point step by step.

Similar to the post on the Kullback Leibler Divergence I created a jupyter notebook with the blog post. You can find it here.

There is also a PDF version of the explanations which has by far the nicest looks (e.g. all the equations are aligned). You can download it here