Better pseudorandom generators via milder pseudorandom restrictions

Publication information:

Gopalan, Parikshit, Raghu Meka, Omer Reingold, Luca Tevisan, and Salil Vadhan. “Better Pseudorandom Generators via Milder Pseudorandom Restrictions”. In Proceedings of the 53rd Annual IEEE Symposium on Foundations of Computer Science (FOCS ‘12), 120-29. IEEE, 2012.

Abstract

Version HistoryFull version posted as ECCC TR12-123 and as arXiv:1210.0049 [cs.CC].


We present an iterative approach to constructing pseudorandom generators, based on the repeated application of mild pseudorandom restrictions. We use this template to construct pseudorandom generators for combinatorial rectangles and read-once \(\mathsf{CNF}\)s and a hitting set generator for width-3 branching programs, all of which achieve near-optimal seed-length even in the low-error regime: We get seed-length \(\tilde{O}(\log(n/\epsilon))\) for error \(\epsilon\). Previously, only constructions with seed-length \(O(log^{3/2}n)\) or \(O(log^2n)\)were known for these classes with error \(\epsilon = 1/\mathrm{poly}(n)\). The (pseudo)random restrictions we use are milder than those typically used for proving circuit lower bounds in that we only set a constant fraction of the bits at a time. While such restrictions do not simplify the functions drastically, we show that they can be derandomized using small-bias spaces.