Research map / Learning theory and algorithms / Matrix and tensor methods
A Network That Learns Strassen Multiplication
Veit Elser
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove one multiplier (rank 6 instead of 7) (ablation) | conservative learning on M2 · M2 (2x2 matrix multiplication tensor) | decomposition error convergence: error fluctuates indefinitely | Hurt |
| use 23 multipliers instead of fewer for 3x3 (ablation) | conservative learning on M3 · M3 (3x3 matrix multiplication tensor) | fraction of runs converging linearly: about 64% of all runs converge | Improved |
| use 22 multipliers instead of 23 (ablation) | conservative learning on M3 · M3 (3x3 matrix multiplication tensor) | convergence behavior: only slow convergence with growing weights | Hurt |
| vary scale of initial weights (ablation) | conservative learning · M2 and M3 | performance | No clear effect |
Closest papers by meaning
- STNet: Spectral Transformation Network for Solving Operator Eigenvalue Problem (NeurIPS 2025 · similarity 0.85)
- Learning to Discover Iterative Spectral Algorithms (NeurIPS 2026 · similarity 0.84)
- Geometry-aware training of factorized layers in tensor Tucker format (NeurIPS 2024 · similarity 0.84)
- The Nuclear Route: Sharp Asymptotics of ERM in Overparameterized Quadratic Networks (NeurIPS 2025 · similarity 0.84)
- Schur's Positive-Definite Network: Deep Learning in the SPD cone with structure (ICLR 2025 · similarity 0.84)
- Learning Polynomial Problems with $SL(2, \mathbb{R})$-Equivariance (ICLR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
