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 ...

“Here we report a deep reinforcement learning approach based on AlphaZero for discovering efficient and provably correct algorithms for the multiplication of arbitrary matrices,” states the paper.

Matrix multiplication is a fundamental operation ... scaling up AlphaTensor to handle even larger tensors would unlock more-efficient algorithms. — Matej Balog and Alhussein Fawzi are at ...

Perfecting that algorithm has been the key to breakthroughs in matrix multiplication efficiency over the past century—even before computers entered the picture. In October 2022, we covered a new ...

Researchers can then break this tensor up into a sum of elementary components, called “rank-1” tensors; each of these will represent a different step in the corresponding matrix multiplication ...

When the player manages to do so, this results in a provably correct matrix multiplication algorithm for any pair of matrices, and its efficiency is captured by the number of steps taken to zero ...

Ever since the Strassen algorithm was published in 1969 ... Today, companies use expensive GPU hardware to boost matrix multiplication efficiency, so any extra speed would be game-changing ...

For AI systems that train on massive computational grids, this efficiency gain translates ... that discovered multiple new matrix multiplication algorithms. One discovery toppled a mathematical ...

The new method, called Linear-Complexity Multiplication (L-Mul), comes close to the results of FPM while using the simpler algorithm ... can be more power efficient, then it seems that we can ...