The algorithm is able to re-discover older matrix multiplication algorithms and improve upon its own to discover newer and faster algorithms. “AlphaTensor is the first AI system for discovering novel, ...

Tensor for matrix multiplication and algorithms: here multiplication of 2 x 2 matrices. Entries equal to 1 are purple, 0 entries are semi-transparent.

The researchers turned the task of discovering new matrix multiplication algorithms into a game dubbed TensorGame. To win the game, an AI system must complete a complex mathematical operation as a ...

A new research paper titled “Discovering faster matrix multiplication algorithms with reinforcement learning” was published by researchers at DeepMind. “Here we report a deep reinforcement learning ...

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. Last ...

Oct 06, 2022 11:20:00 The strongest shogi AI reaches new ground, DeepMind's AI 'AlphaTensor' succeeds in improving the matrix multiplication algorithm that has been stagnant for over 50 years ...

It is compatible across many different compilers, languages, operating systems, linking, and threading models. In particular, the Intel MKL DGEMM function for matrix-matrix multiplication is highly ...