Google has developed a new machine learning "monster" called JAX, which is Autograd and XLA, brought together for high-performance machine learning research.
With its updated version of Autograd, JAX can automatically differentiate native Python and NumPy functions. It can differentiate through loops, branches, recursion, and closures, and it can take derivatives of derivatives of derivatives. It supports reverse-mode differentiation (a.k.a. backpropagation) via grad as well as forward-mode differentiation, and the two can be composed arbitrarily to any order. What’s new is that JAX uses XLA to compile and run your NumPy programs on GPUs and TPUs. See more Google JAX.
-
- Active Topics
-
-
- by Eli 8 hours ago Inaugural Lecture by Prof. Mukandala: Dola, Soko na Kushindwa kwa Taasisi View the latest post Replies 2 Views 62
- by Eli 1 day ago All in One: YouTube, TED, X, Facebook and Instagram Reels, Videos, Images and Text Posts View the latest post Replies 332 Views 41531
- by Eli 1 day ago Iran's President Ebrahim Raisi Aged 63 Dies in a Helicopter Crash View the latest post Replies 3 Views 66
- by Eli 1 day ago Re: What is in Your Mind? View the latest post Replies 717 Views 308348
- by Eli 4 days ago PySpark for Large Data Processing View the latest post Replies 2 Views 8189
- by Eli 4 days ago Online Bible View the latest post Replies 3 Views 23334
- by Eli 4 days ago Generating SSH Key and Adding it to the ssh-agent for Authentication on GitHub View the latest post Replies 1 Views 515
- by Eli 1 week ago Russia Invades Ukraine View the latest post Replies 663 Views 242239
- by Eli 2 weeks ago President Museveni's Speech During International Development Association (IDA) Summit View the latest post Replies 1 Views 509
- by Eli 2 weeks ago From Simple Linear Regression Analysis to Covariance & Correlation to Independent Determinant, and R-Squared View the latest post Replies 11 Views 25154
-
Google JAX: Autograd and XLA
-
- Similar Topics
- Replies
- Views
- Last post
-
- 0 Replies
- 759 Views
-
Last post by Eli
-
- Information
-
Who is online
Users browsing this forum: No registered users and 5 guests