English

Pseudorandom Generators for Width-3 Branching Programs

Computational Complexity 2018-06-13 v1

Abstract

We construct pseudorandom generators of seed length O~(log(n)log(1/ϵ))\tilde{O}(\log(n)\cdot \log(1/\epsilon)) that ϵ\epsilon-fool ordered read-once branching programs (ROBPs) of width 33 and length nn. For unordered ROBPs, we construct pseudorandom generators with seed length O~(log(n)poly(1/ϵ))\tilde{O}(\log(n) \cdot \mathrm{poly}(1/\epsilon)). This is the first improvement for pseudorandom generators fooling width 33 ROBPs since the work of Nisan [Combinatorica, 1992]. Our constructions are based on the `iterated milder restrictions' approach of Gopalan et al. [FOCS, 2012] (which further extends the Ajtai-Wigderson framework [FOCS, 1985]), combined with the INW-generator [STOC, 1994] at the last step (as analyzed by Braverman et al. [SICOMP, 2014]). For the unordered case, we combine iterated milder restrictions with the generator of Chattopadhyay et al. [CCC, 2018]. Two conceptual ideas that play an important role in our analysis are: (1) A relabeling technique allowing us to analyze a relabeled version of the given branching program, which turns out to be much easier. (2) Treating the number of colliding layers in a branching program as a progress measure and showing that it reduces significantly under pseudorandom restrictions. In addition, we achieve nearly optimal seed-length O~(log(n/ϵ))\tilde{O}(\log(n/\epsilon)) for the classes of: (1) read-once polynomials on nn variables, (2) locally-monotone ROBPs of length nn and width 33 (generalizing read-once CNFs and DNFs), and (3) constant-width ROBPs of length nn having a layer of width 22 in every consecutive polylog(n)\mathrm{poly}\log(n) layers.

Keywords

Cite

@article{arxiv.1806.04256,
  title  = {Pseudorandom Generators for Width-3 Branching Programs},
  author = {Raghu Meka and Omer Reingold and Avishay Tal},
  journal= {arXiv preprint arXiv:1806.04256},
  year   = {2018}
}

Comments

51 pages