Deep Generative Models and Inverse Problems
This is a talk by Alexandros Dimakis given in April 2020 for a seminar at the Institute for Advanced Study seminar on theoretical machine learning.
So he organizes the world of modeling high dimensionality distributions into 3 categories.
1: Sparsity (so think wavelets).
2: Conditional Independence (so think markov chains, factor graphs, bayes nets used in things like channel coding).
3: Deep Generative Models (so think passing random noise through a learnable differentiable function (like GANS, VAEs, etc).