Faster Algorithms for Structured Matrix Multiplication via Flip Graph Search

Abstract

We give explicit low-rank bilinear non-commutative schemes for multiplying structured $n \times n$ matrices with $2 \leq n \leq 5$, which serve as building blocks for recursive algorithms with improved multiplicative factors in asymptotic complexity. Our schemes are discovered over $\mathbb{F}_2$ or $\mathbb{F}_3$ and lifted to $\mathbb{Z}$ or $\mathbb{Q}$. Using a flip graph search over tensor decompositions, we derive schemes for general, upper-triangular, lower-triangular, symmetric, and skew-symmetric inputs, as well as products of a structured matrix with its transpose. These schemes improve asymptotic constants for 13 of 15 structured formats. In particular, we obtain $4 \times 4$ rank-34 schemes for both multiplying a general matrix by its transpose and an upper-triangular matrix by a general matrix, improving the asymptotic factor from 8/13 (0.615) to 22/37 (0.595). Additionally, using $\mathbb{F}_3$ flip graphs, we discover schemes over $\mathbb{Q}$ that fundamentally require the inverse of 2, including a $2 \times 2$ symmetric-symmetric multiplication of rank 5 and a $3 \times 3$ skew-symmetric-general multiplication of rank 14 (improving upon AlphaTensor's 15).