Skip to content

Author

V. Grishin

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

Quantitative tiling stability from quadratic discrepancy in Hamming spaces

Quadratic ball discrepancy defines an energy on codes in finite Hamming spaces. At perfect-code parameters, its exact minimizers are the perfect codes. We fix the alphabet size, length, and code cardinality and compare all codes with these parameters. We prove tiling-defect stability: excess discrepancy above the perfect-code benchmark controls the squared deviation of the distinguished ball-covering multiplicity from one. For one-error parameters satisfying sphere-packing and Lloyd integrality, the lower coefficient is $\kappa_{n,q}/q^2\geq1$. The uniform floor one is sharp, while the certified parameter-dependent coefficient can be much larger. An explicit parameter-dependent upper estimate is also available, and the two certified coefficients can be far apart. For any two-error parameter pair with $n\geq5$ satisfying sphere-packing divisibility and having two distinct integral Lloyd roots in the Hamming weight range, we obtain an explicit positive coefficient without assuming that a perfect code exists. For alphabets of size at least four, this conditional coefficient has a closed form and fixed-alphabet asymptotics. Direct certificates for the repetition and Golay families, combined with perfect-code classification, give tiling-defect stability for every nontrivial perfect code. Here stability concerns the ball-covering multiplicity profile, not symmetric-difference proximity to a particular perfect code. The defect is also a normalized chi-square smoothing error under uniform ball noise, so excess discrepancy controls holes, overlaps, defective ambient points, total variation, and R\'enyi divergence from uniformity of the ball-noise output. Competing codes need not be linear or satisfy a distance or error-correction constraint.

V. Grishin, Aryeh Lev Zabokritskiy · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.