JAX

Summary:
JAX is Autograd and XLA, brought together for high-performance machine learning research.

URL: https://github.com/google/jax

Developers:

• "Who is in the JAX group at google?" Bing w/ ChatGPT responds with a possibly-incomplete list. Matt Johnson: Tech Lead for JAX
• Roy Frostig: Software Engineer for JAX
• Dougal Maclaurin: Software Engineer for JAX
• James Bradbury: Software Engineer for JAX
• Skye Wanderman-Milne: Software Engineer for JAX
• Peter Hawkins: Software Engineer for JAX
• Alex Wiltschko: Research Scientist for JAX
• Sam Schoenholz: Research Scientist for JAX
• Ekin Dogus Cubuk: Research Scientist for JAX
• George Dahl: Research Scientist for JAX

Mode:	Forward Reverse
Method:	Source transformation Operator overloading
Supported Language:	Python

Features:
Google is moving most of its machine learning and neural network systems over to JAX.

Licensing: open source

Entries in our publication database that actually use JAX in the numerical experiments:  3

The following diagram shows these entries versus the year of the publication.

10+ #Entries 0	1	2
	'20	'21
	Year