ETE 2015 – Soumith Chintala – The deep learning revolution: rethinking machine learning pipelines

by
Tags: , , ,

Download (PDF, 1.91MB)

Abstract:

In the last decade, a class of machine learning algorithms popularly know as “deep learning” have produced state-of-the-art results on a wide variety of domains, including image recognition, speech recognition, natural language processing, genome sequencing, financial data among others.

Facebook, Google, Twitter, NVIDIA, Apple and many other tech giants are investing heavily into deep learning, which is powered by Graphics Processing Units to accelerate computation and scale up to larger and larger deep learning models.

What is deep learning? Why has it become so popular so quickly? How can one fit deep learning into existing pipelines?

About Soumith:

I am a Research Engineer at Facebook AI Research. Prior to joining Facebook in August 2014, I worked at MuseAmi, where I built deep learning models for music and vision targeted at mobile devices. I hold a Masters in CS from NYU, and spent time in Yann’s lab building deep learning models for pedestrian detection, natural image OCR, depth-images among others.