DeepMind x UCL | Deep Learning Lectures | 11/12 | Modern Latent Variable Models

This lecture, by DeepMind Research Scientist Andriy Mnih, explores latent variable models, a powerful and flexible framework for generative modelling. After introducing this framework along with the concept of inference, which is central to it, Andriy focuses on two types of modern latent variable models: invertible models and intractable models. Special emphasis is placed on understanding variational inference as a key to training intractable latent variable models.
Note this lecture was originally advertised as lecture 9.
Download the slides here:
storage.googleapis.com/deepmi...
Find out more about how DeepMind increases access to science here:
deepmind.com/about#access_to_...
Speak Bio:
Andriy Mnih is a Research Scientist at DeepMind. He works on generative modelling, representation learning, variational inference, and gradient estimation for stochastic computation graphs. He did his PhD on learning representations of discrete data at the University of Toronto, where he was advised by Geoff Hinton. Prior to joining DeepMind, Andriy was a post-doctoral researcher at the Gatsby Unit, University College London, working with Yee Whye Teh.
About the lecture series:
The Deep Learning Lecture Series is a collaboration between DeepMind and the UCL Centre for Artificial Intelligence. Over the past decade, Deep Learning has evolved as the leading artificial intelligence paradigm providing us with the ability to learn complex functions from raw data at unprecedented accuracy and scale. Deep Learning has been applied to problems in object recognition, speech recognition, speech synthesis, forecasting, scientific computing, control and many more. The resulting applications are touching all of our lives in areas such as healthcare and medical research, human-computer interaction, communication, transport, conservation, manufacturing and many other fields of human endeavour. In recognition of this huge impact, the 2019 Turing Award, the highest honour in computing, was awarded to pioneers of Deep Learning.
In this lecture series, research scientists from leading AI research lab, DeepMind, deliver 12 lectures on an exciting selection of topics in Deep Learning, ranging from the fundamentals of training neural networks via advanced ideas around memory, attention, and generative modelling to the important topic of responsible innovation.

[@leixun]

*DeepMind x UCL | Deep Learning Lectures | 11/12 | Modern Latent Variable Models*
*My takeaways:*
*1. Lecture outline **0:38*
*2. Generative modeling **1:45*
2.1 Introduction of generative models 1:50
2.2 Progress in generative models 6:30
2.3 Types of generative models 8:00
*3. Latent variable models & inference **15:11*
*4. Invertible models & exact inference **30:15*
*5. Variational inference (VI) **41:47*
*6. Gradient estimation in VI **1:10:25*
*7. Variational autoencoders **1:22:15*
*8. Conclusion **1:27:04*

[@user-or7ji5hv8y]

best presentation on generative approach so far.

[@SempoiGiler]

Thank you, really grateful for these kinds of educational videos, especially from a country where AI research can be considered non-existence. While I have a big desire to learn it. The knowledge shared is so precious to me. Thanks thanks thanks.

[@user-or7ji5hv8y]

thank you so much for posting these videos!

[@youvenzful]

Big thanks for this incredible overview of latent variable models in a such short time presentation! Indeed, some additional material or literature recommendations on the subject could have been helpful.

[@learnml7034]

Great presentation - very clear!

[@patricknnamdi2203]

This was great, thanks!

[@lukn4100]

Great lecture and big thanks to DeepMind for sharing this great content.

[@colevfrank]

Superb lecture--very clear explanation of variational autoencoders and the associated tradeoffs between ease of inference and modeling flexibility/power

[@bryanbosire]

good work expounding on applications of statistical inference in generative models

[@user-or7ji5hv8y]

How about conjugate prior for tractability?

[@justiny.8365]

any recommonded literature to learn more about this?

[@bingeltube]

Unfortunately, Mnih provides very few references in video and slides

[@user-or7ji5hv8y]

What is a factorial prior?

[@freemind.d2714]

The slides seem missing, please fix the link!

[@tunestar]

Love the topic!! On to the lesson...

[@tunestar]

Ok, it sucked! Why? Too much theory and math, very few practical examples.

[@user-ls7mr4rq5x]

I never tease people like you