The ' Strassen's algorithm ' announced in 1969 is widely known as an algorithm for calculating the matrix product at high speed.

For such enormous numbers, even Karatsuba's algorithm is too slow. A real breakthrough came in 1971 with the work of the German mathematicians Arnold Schönhage and Volker Strassen.

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

What do encrypted messages, recognizing speech commands and running simulations to predict the weather have in common? They all rely on matrix multiplication for accurate calculations. DeepMind, an ...

So if you do 32x32 with the naive algorithm, and 64x64 with strassen, you save ONE n^3 matrix multiplication and lose EIGHTEEN n^2 matrix additions with n = 32 for a massive overall saving.

Improving Matrix operation algorithms is great, but we're well into the long tail of diminishing returns.