Asymptotic Rank Speedup Theorems, Revisited
Josh Alman, Baitian Li

TL;DR
This paper develops a unified framework for improving asymptotic tensor rank bounds, with applications to matrix multiplication complexity and tensor analysis.
Contribution
It generalizes classical speedup theorems, enabling quantitative improvements in asymptotic tensor rank bounds beyond previous results.
Findings
Asymptotic rank of small Coppersmith-Winograd tensor is less than its border rank.
Improved upper bound below d^{2ω/3} for asymptotic rank of any d×d×d tensor.
Introduces Strassen calculus for analyzing degenerations with nontrivial direct sums.
Abstract
Motivated by fast matrix multiplication and recent connections between asymptotic tensor rank and fine-grained complexity, we revisit classical tools from the matrix multiplication literature and develop a framework for obtaining improved asymptotic rank upper bounds for tensors beyond matrix multiplication. In the 1980s, Coppersmith-Winograd and Strassen discovered a series of speedup theorems for asymptotic rank: in certain regimes, one can extract additional terms from a border rank upper bound on a tensor , and then use these terms to obtain an improved asymptotic rank of . We establish general speedup theorems that subsume these results and enable quantitative improvements. Two representative applications are: (1) The asymptotic rank of the small Coppersmith-Winograd tensor is less than its border rank. For instance, we prove the asymptotic rank of…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
