November 1, 2023

Spring 2024 Course: ECE 69500GM Generative Models

Spring 2024 Course: ECE 69500GM Generative Models
 
ECE695 (Spring 2024): "Inference and Learning in Generative Models"
 
An introduction to modern generative models like diffusion models, variational autoencoders, normalizing flows, and energy-based models, with a focus on derivations from the perspective of statistical learning theory.  We build up from the basics, starting with probabilistic graphical models, which provide the framework for many of the ideas in the class.
 
Generative models
  •     have had an enormous impact on public life in the last year (Stable diffusion, DALL-E, Imagen; GPT, PixelRNN, Wavenet; etc.) and this is likely to continue.
  •     are an extremely active research area.
  •     have applications in numerous domains and research problems
Prerequisites:
    linear algebra, multivariable calculus, basic probability/statistics (no course pre-reqs)
 
Please direct any questions to Prof. J.G. Makin (jgmakin@purdue.edu / BHEE 330).