Matrix multiplication is a fundamental operation in machine learning, and is one of the most time-consuming, due to the extensive use of multiply-add instructions.

High-performance matrix multiplication remains a cornerstone of numerical computing, underpinning a wide array of applications from scientific simulations to machine learning.

Can artificial intelligence (AI) create its own algorithms to speed up matrix multiplication, one of machine learning’s most fundamental tasks? Today, in a paper published in Nature, DeepMind ...

DeepMind breaks 50-year math record using AI; new record falls a week later AlphaTensor discovers better algorithms for matrix math, inspiring another improvement from afar.