English

The smallest grammar problem revisited

Data Structures and Algorithms 2019-08-20 v1

Abstract

In a seminal paper of Charikar et al. on the smallest grammar problem, the authors derive upper and lower bounds on the approximation ratios for several grammar-based compressors, but in all cases there is a gap between the lower and upper bound. Here the gaps for LZ78\mathsf{LZ78} and BISECTION\mathsf{BISECTION} are closed by showing that the approximation ratio of LZ78\mathsf{LZ78} is Θ((n/logn)2/3)\Theta( (n/\log n)^{2/3}), whereas the approximation ratio of BISECTION\mathsf{BISECTION} is Θ(n/logn)\Theta(\sqrt{n/\log n}). In addition, the lower bound for RePair\mathsf{RePair} is improved from Ω(logn)\Omega(\sqrt{\log n}) to Ω(logn/loglogn)\Omega(\log n/\log\log n). Finally, results of Arpe and Reischuk relating grammar-based compression for arbitrary alphabets and binary alphabets are improved.

Keywords

Cite

@article{arxiv.1908.06428,
  title  = {The smallest grammar problem revisited},
  author = {Hideo Bannai and Momoko Hirayama and Danny Hucke and Shunsuke Inenaga and Artur Jez and Markus Lohrey and Carl Philipp Reh},
  journal= {arXiv preprint arXiv:1908.06428},
  year   = {2019}
}

Comments

A short version of this paper appeared in the Proceedings of SPIRE 2016. This work has been supported by the DFG research project LO 748/10-1 (QUANT-KOMP)

R2 v1 2026-06-23T10:50:06.563Z