Remind me of an article about the opimization of matrix calculus with IA, that eventually shown better algorithm for specific size of matrix, than the one found decades ago and proven to be the best one.
> AlphaTensor discovered algorithms that outperform the state-of-the-art complexity for many matrix sizes. Particularly relevant is the case of 4 × 4 matrices in a finite field, where AlphaTensor’s algorithm improves on Strassen’s two-level algorithm for the first time, to our knowledge, since its discovery 50 years ago.
https://towardsdatascience.com/how-deepmind-discovered-new-w...