Orbit compression and asymptotic contractivity for symmetric Bohnenblust--Hille inequalities
arXiv:2608.13753
2026
Architecture
2 ideas extracted · analyzed Sep 1, 2026
What the math gives to ML
The paper identifies multinomial orbit multiplicities as the quantity controlling norm distortion when a symmetric multilinear form is represented by its diagonal polynomial. Its transferable asset is an exact orbit-weighted coefficient representation together with asymptotic contractivity: for complex symmetric degree-m forms, the norm ratio approaches one at rate O(1/m), with a related anisotropic bound of O(1/sqrt(m)). This suggests multiplicity-aware parameterization and regularization for symmetric tensor or polynomial neural layers, avoiding the coefficient and gradient distortions caused by treating every monomial orbit as unweighted. The low/high-support cutoff also provides a principled sparsification strategy for high-degree interaction features.
Ideas from this paper
△ Mechanism confirmed, baseline not beaten
2026
Replace a dense degree-m tensor interaction layer by a symmetric orbit-parameterized layer with one parameter per exponent vector and explicit multinomial scaling. This preserves the contribution of all ordered tensor entries represented by one orbit, while reducing parameter count and avoiding the amplitude distortion of unweighted monomial compression.
Useful7/10
Difficulty4/10
Novelty7/10
Unverified
2026
Separate low-support monomials, which involve only a few distinct input coordinates, from high-support monomials in a high-degree symmetric interaction layer. Compute the low-support orbit features exactly and prune, sample, or factorize the high-support tail, using the paper's cutoff scale as the initial sparsity rule.
Useful6/10
Difficulty5/10
Novelty8/10