My knowledge about Expectation Maximization (EM) is limited, from my understanding, EM is just an algorithm to do optimization. It works well when we have some hidden / latent variables, such as Hidden Markov model or Mixture of Gaussians.
On the other hand, I learned many other continuous optimization algorithms such as gradient decent, newton's method, etc.
So what is the relationship between EM and these continuous optimization algorithms? And is that possible for anyone to give me an example of using EM in a quadratic optimization
$$ \min \mathbf x^\top A \mathbf x $$
