Composing Graphical Models With Neural Networks with David Duvenaud

800 800 The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

In this episode, we hear from David Duvenaud, assistant professor in the Computer Science and Statistics departments at the University of Toronto. David joined me after his talk at the Deep Learning Summit on “Composing Graphical Models With Neural Networks for Structured Representations and Fast Inference.”

In our conversation, we discuss the generalized modeling and inference framework that David and his team have created, which combines the strengths of both probabilistic graphical models and deep learning methods. He gives us a walkthrough of his use case which is to automatically segment and categorize mouse behavior from raw video, and we discuss how the framework is applied here and for other use cases. We also discuss some of the differences between the frequentist and bayesian statistical approaches.

This show is part of a series of shows recorded at the RE•WORK Deep Learning Summit in Montreal back in October. This was a great event and, in fact, their next event, the Deep Learning Summit San Francisco is right around the corner on January 25th and 26th, and will feature more leading researchers and technologists like the ones you’ll hear here on the show this week, including Ian Goodfellow of Google Brain, Daphne Koller of Calico Labs, and more! Definitely check it out and use the code TWIMLAI for 20% off of registration.

Giveaway Update!

Thanks to everyone who took the time to enter our #TWIML1MIL listener giveaway! We sent out an email to entrants a few days ago, so please be on the lookout for that. If you haven’t heard from us yet, please reach out to us at so that we can get you your swag!

TWIML Online Meetup

The details for our January Meetup are set! Tuesday, January 16, we will be joined by former TWIML guest and Microsoft Researcher Timnit Gebru. Timnit joined us a few weeks ago to discuss her recently released, and much acclaimed paper, “Using deep learning and Google Street View to estimate the demographic makeup of neighborhoods across the United States”, and I’m excited that she’s be joining us to discuss the paper, and the pipeline she used to identify 22 million cards in 50 million Google Street View images, in more detail. I’m anticipating a lively discussion segment, in which we’ll be exploring your AI resolutions & predictions for 2018. For links to the paper, or to register for the meetup, or to check out previous meetups, visit

About David

Mentioned in the Interview

Leave a Reply

Your email address will not be published.