This week on the podcast we’re featuring a series of conversations from the NIPS conference in Long Beach, California. This was my first time at NIPS, and I had a great time there: I attended a bunch of talks and of course learned a ton, I organized an impromptu roundtable on Building AI Products, and met a bunch of great people, including some former TWiML Talk guests. I’ll be sharing a bit more about my experience at NIPS via my newsletter, which you should take a second right now to subscribe to at twimlai.com/newsletter.

Thanks to our Sponsor

I’d like to thank our friends over at Intel Nervana for their sponsorship of this podcast and our NIPS series. While Intel was very active at NIPS, with a bunch of workshops, demonstrations and poster sessions, their big news at NIPS was the first public viewing of the Intel Nervana™ Neural Network Processor, or NNP. The goal of the NNP architecture is to provide the flexibility needed to support deep learning primitives while making the core hardware components as efficient as possible, giving neural network designers powerful tools for solving larger and more difficult problems while minimizing data movement and maximizing data re-use. To learn more about Intel’s AI Products Group and the Intel Nervana NNP, visit IntelNervana.com.

TWiML Presents: NIPS 2017

TWiML Talk #88Using Deep Learning and Google Street View to Estimate DemographicsTimnit Gebru, Microsoft Research

TWiML Talk #89AI at the NASA Frontier Development LabSara Jennings, NASA Frontier Development Lab
Timothy Seabrook, NASA Frontier Development Lab
Andres Rodriguez, Intel

TWiML Talk #90Geometric Deep LearningJoan Bruna, NYU
Michael Bronstein, Università della Svizzera Italiana

TWiML Talk #91Philosophy of IntelligenceMatthew Crosby, Imperial College

TWiML Talk #92Learning State RepresentationsYael Niv, Princeton