English

R\'enyi divergence guarantees for hashing with linear codes

Information Theory 2025-06-06 v2 math.IT

Abstract

We consider the problem of distilling uniform random bits from an unknown source with a given pp-entropy using linear hashing. As our main result, we estimate the expected pp-divergence from the uniform distribution over the ensemble of random linear codes for all integer p2p\ge 2. The proof relies on analyzing how additive noise, determined by a random element of the code from the ensemble, acts on the source distribution. This action leads to the transformation of the source distribution into an approximately uniform one, a process commonly referred to as distribution smoothing. We also show that hashing with Reed-Muller matrices reaches intrinsic randomness of memoryless Bernoulli sources in the lpl_p sense for all integer p2p\ge 2.

Keywords

Cite

@article{arxiv.2405.04406,
  title  = {R\'enyi divergence guarantees for hashing with linear codes},
  author = {Madhura Pathegama and Alexander Barg},
  journal= {arXiv preprint arXiv:2405.04406},
  year   = {2025}
}

Comments

Minor changes from v1. Final version, to appear in IEEE Transactions on Information Theory

R2 v1 2026-06-28T16:19:38.600Z