Course Description

Deep generative models aim to combine the interpretable representations and quantified uncertainty offered by probabilistic models, with the flexibility and scalable learning of deep neural networks. This research area -- which includes variational autoencoders, generative adversarial networks, and more -- is one of the most exciting and rapidly evolving fields of statistical machine learning. This seminar course will explore the cutting edge of deep generative models. This is a PhD-level seminar course, and will involve a major project by the student (or group of two students). Students have substantial freedom in choosing these projects, from creating novel research-grade methods, to contributing to open source machine learning projects, to analyzing data of interest, to exploring a theoretical topic. Projects that connect to the student's ongoing research agenda are particularly welcome. Proposals and reports will be done in the format and style of top machine learning conferences, with the intention that some projects will be developed into papers at those venues.

Project materials

Presentation materials

Administrative

  • Instructor: John P. Cunningham
  • Lecture: Wed 2:10PM–4:00PM, in Stat 1025
  • OH (pres): Mon 10:00AM–11:00AM, in Stat 1026
  • OH (proj): Wed 4:00PM–5:00PM, in Stat 1026

Prerequisites

Course Content

Date Reading Slides Presenter(s)
23 Jan Intro and basics pdf John
30 Jan Adding structure part I: dynamics pdf John
06 Feb Adding structure part II: discrete 1 pdf Elliott and Hooshmand
13 Feb Adding structure part II: discrete 2 pdf Yixin and Gabriel
20 Feb More reparameterizations and control variates pdf Andy Miller
27 Feb Improving the ELBO: better bounds pdf Luhuan and Amin
06 Mar Improving the ELBO: disentangling latent representations pdf Benjamin and Andrew
13 Mar Improving the ELBO: disentangling part 2 pdf Sean and Chris
20 Mar Even more gradient variance (spring break...) no slides, no lecture no one
27 Mar GANsition pdf Peter and Kelly
03 April Wasserstein GAN pdf Taiga and Ding
10 April Improving GAN pdf Chengliang and Yunhao
17 April GAN more closely: theory, numerics, empiricism pdf Zhen and Wenda
24 April GAN in NLP pdf Yao and Tom E.
31 April GAN in CV to come Tom B. and Fan