What is JAX?

JAX is a high performance numerical computing framework that brings together differentiation to Python code (Autograd) and Accelerated Linear Algebra (XLA) that compiles to low level, high performing code on accelerators, such as GPUs and TPUS. In this video, Laurence Moroney discusses what JAX is, what it does, and why it’s important in Google’s research efforts.

Resources:
JAX reference docs → https://goo.gle/3IE4UQ7
GitHub repository→ https://goo.gle/3IQaK0O

Chapters:
0:00 - Intro
0:58 - Concepts to consider in ML frameworks
1:30 - What is the idea behind JAX?
2:33 - What are the main benefits of using JAX?
3:29 - Leave us questions in the comments!

Watch more episodes of JAX Foundations → https://goo.gle/JAXFoundations
Subscribe to the Google Research Channel → https://goo.gle/GoogleResearch