Improved Pseudorandom Generators for π π’^0 Circuits
We show a new PRG construction fooling depth-d, size-m π π’^0 circuits within error Ξ΅, which has seed length O(log^d-1(m)log(m/Ξ΅)loglog(m)). Our PRG improves on previous work (Trevisan and Xue 2013, Servedio and Tan 2019, Kelley 2021) from various aspects. It has optimal dependence on 1/Ξ΅ and is only one βloglog(m)β away from the lower bound barrier. For the case of d=2, the seed length tightly matches the best-known PRG for CNFs (De et al. 2010, Tal 2017). There are two technical ingredients behind our new result; both of them might be of independent interest. First, we use a partitioning-based approach to construct PRGs based on restriction lemmas for π π’^0, which follows and extends the seminal work of (Ajtai and Wigderson 1989). Second, improving and extending prior works (Trevisan and Xue 2013, Servedio and Tan 2019, Kelley 2021), we prove a full derandomization of the powerful multi-switching lemma for a family of DNFs (HΓ₯stad 2014).
READ FULL TEXT