Variational Autoencoders

A lecture that discusses variational autoencoders. We discuss generative models, plain autoencoders, the variational lower bound and evidence lower bound, variational autoencoder architecture, and stochastic optimization of the variational lower bound.
This lecture is from Northeastern University's CS 7150 Summer 2020 class on Deep Learning, taught by Paul Hand.
The notes are available at: khoury.northeastern.edu/home/h...
References:
Kingma and Welling 2019:
Kingma, Diederik P., and Max Welling. "An Introduction to Variational Autoencoders." Foundations and Trends® in Machine Learning 12, no. 4 (2019): 307-392. arxiv.org/abs/1906.02691
Kingma and Welling 2014:
Kingma, Diederik P., and Max Welling. "Auto-encoding variational bayes." arXiv preprint arXiv:1312.6114 (2013).
Razavi et al. 2019:
Razavi, Ali, Aaron van den Oord, and Oriol Vinyals. "Generating diverse high-fidelity images with VQ-VAE-2." In Advances in Neural Information Processing Systems, pp. 14866-14876. 2019.

[@gorgolyt]

Best explanation on RUclips. Exactly what I was looking for. Thorough, logical, intuitive.

[@yurigansmith]

Very good presentation. Thanks a lot!

[@wilsonlwtan3975]

This is a gem. Finally, someone that is able to give concise teaching well! Thank you!

[@bluestar2253]

One of the best explanations on VAE on YT. Thank you and keep up the good work!

[@amirhosseinramazani757]

I enjoyed your explanation. I needed something like this video to get a little deeper into the theory of the VAEs. Thank you!

[@pietrocestola7856]

Clear, concise and very accurate. Thank you so much for sharing with us this wonderful explanation.

[@sucramgnat8157]

Thank you so much for your lecture. You truly have a talent for teaching!

[@user-or7ji5hv8y]

wow, this is so well explained.

[@Procuste34iOSh]

thank you so much. so underrated

[@trongduong1047]

very nice explanation!

[@slemanbisharat6390]

Thank you excellent explanation!!

[@MeowlaMars]

This is clear and awesome

[@gomctigger4439]

Hi @Paul Hand,
thank you for the lecture.
What is the intuition behind using q(z|x) in the expectation or the expectation at all? I see that it makes sense mathematically, but how would one get the idea?
In contrast, there is a derivation of the ELBO via importance sampling and then applying Jensen Inequality or via the optimal sampler.

[@sahhaf1234]

How do we know that p(x|z) is normally distributed?? Do we just assume it?
x|z is just a neural network and I dont see any reason for p(x|z) to distribute normally. Actually, the relation between x and z must be deterministic.

[@hubertnguyen8855]

Very nice and comprehensive lecture. Thanks

[@yurigansmith]

Training of generative models start here: 6:26

[@oFabianLoL]

I don't understand what phi and theta mean. "the parameters of the model", does that mean the weights of the neural network? or the parameters of the distribution, eg if it is gaussian, the parameters correspond to a mu and sigma.
I appreciate if anyone can clarify, thank you!

[@ThatQCboy]

parameters of the model. we use MLE principles to find the optimal phi and theta

[@doyney]

I'm pretty sure phi and theta represent the parameters in terms of weights and biases in the encoder/decoder neural networks.

[@madhusudanverma6564]

24:48, how maximizing vlb will roughly maximize p(x) because, since x is given p(x) should be constant.

[@josephpalermo8898]

p(x) is actually parameterized therefore it's not constant

[@maximmaximov4147]

It would be really perfect if someone started giving some examples on each step since we are talking about real things that exist in the world. Each step has its meaning and intention and is made to overcome challenges or obstacles that come up on the way. I want to know what we are doing and what is the purpose. And what is gonna happen if we wouldn't do it this way. I cannot find anything non abstract, I need examples to put my imagination on. It is clear and good only if you have prior knowledge of the things being discussed. Otherwise there are million ways to interpret things and even more to get lost

[@maximmaximov4147]

At 11:00 it seems like if we are talking about pictures the formula written in blue should generate an image with real random noise which doesn't make sense. It should have been done differently like is said in other articles so that random distributions of different images (sets of parameters or pixels) overlap. So that it is not purely random noise which is not we're trying to reach