Getting started with TensorFlow

Tuesday, June 19
11:50 AM - 12:30 PM
Grand Ballroom 220A

In this talk, we'll cover the basics of TensorFlow and will show how Google pushes the boundaries of machine learning using TensorFlow and cloud. We'll look at some of the latest models we've been working on and some of the new APIs in TensorFlow such as:, tf.keras, tf.estimators, etc. We will also do a deep dive into how the Tensor Processing Unit (TPU) ASIC chip that uses systolic arrays works and how you can train very large model effectively through distribution.

But getting started with machine learning today does not necessarily require deep theoretical expertise on how to train models . There are actually quite a lot of cloud services that do the theoretical heavy lifting for you, so we will look at some of those. Finally, we will see how you can use TensorFlow Lite to deploy models to mobile devices. If you want to get started with machine learning on TensorFlow, this session is for you.


Magnus Hyttsten
Developer Advocate
Google Inc
Magnus Hyttsten is a Developer Advocate for TensorFlow @ Google. He works on developing the TensorFlow product, is a developer fanatic, and an appreciated speaker at major industry events such as Google I/O, The AI Summit, AI Conference, ODSC, and MWC on machine learning and mobile development . Right now, he is focusing on Reinforcement Learning models, as well as making model inference effective on Mobile.