# Marginalization When thinking about marginalization, think about summing along one dimension, adding up the results, and then normalizing. Below we are marginalizing out the $x_2$ dimension so we end up only with a probability distribution of $x_1$.  There is a great widget to explore this in the blog post: [A Visual Exploration of Gaussian Processes](https://distill.pub/2019/visual-exploration-gaussian-processes/). --- Date: 20240120 Links to: [Einsum](Einsum.md) Tags: References: * []()