JAX is a system for high-performance machine-learning research. It offers the familiarity of Python NumPy together with hardware acceleration. JAX enables the definition and composition of user-wielded function transformations useful for machine-learning programs. These transformations include automatic differentiation, automatic batching, end-to-end compilation (via XLA), parallelizing over multiple accelerators, and more. Composing these transformations is the key to JAX's power and simplicity.

AMD is finally making its first moves into deep learning. Google will start equipping its sprawling data center infrastructure with AMD's graphics processing units (or GPUs) to accelerate deep learning applications, AMD announced on Tuesday at a supercomputing conference in Salt Lake City, Utah. GPUs, typically used for generating the latest in gaming graphics, have been booming in deep learning, which is a flavor of artificial intelligence where the computer teaches itself how to do certain tasks. GPUs have caught on because of their capabilities in "parallel computing," a technique that involves multiple calculations happening simultaneously. That makes GPUs much faster at running deep learning neural nets than more generalized processors.