Pseudorandomness and Fourier-Growth Bounds for Width-3 Branching Programs

Pseudorandomness and Fourier-Growth Bounds for Width-3 Branching Programs
复制标题

宽度 3 分支程序的伪随机性和傅立叶增长界

DOI:
10.4086/toc.2017.v013a012
复制
发表时间:
2014
期刊:
Electron. Colloquium Comput. Complex.
影响因子:
--
通讯作者:
Andrew Wan
Andrew Wan
中科院分区:
--
文献类型:
--
作者:
T. Steinke;Salil P. Vadhan;Andrew Wan

文献摘要

被引文献

相似文献

我们提出了一个明确的伪和生成器,用于遗忘,读取,宽度3分支程序,可以按任何顺序读取其输入位。发电机的种子长度O〜(log^3 n)。 该模型的先前最著名的种子长度是n^{1/2+o(1)},这是由于不受欢迎的,Meka和Zuckerman(focs'12)。我们的工作概括了Reingold,Steinke和Vadhan(Random'13)的最新结果,用于排列分支程序。我们的发电机的主要技术新颖性是宽度3的傅立叶生长,即遗忘的,一开始的分支程序。具体来说,我们表明,对于任何f:{0,1}^n-> {0,1},由这样的分支程序计算出来,在[n]中计算的k,sum_ {| s | = k} | hat {f} (S)| <n^2 *(o(\ log n))^k, 其中f(x)= sum_s hat {f}(s)(-1)^是Z_2^ n的标准傅立叶变换。傅立叶生长的基本O(log n)紧密到log log n的因素。
We present an explicit pseudorandom generator for oblivious, read-once, width-3 branching programs, which can read their input bits in any order. The generator has seed length O~( log^3 n ). The previously best known seed length for this model is n^{1/2+o(1)} due to Impagliazzo, Meka, and Zuckerman (FOCS'12). Our work generalizes a recent result of Reingold, Steinke, and Vadhan (RANDOM'13) for permutation branching programs. The main technical novelty underlying our generator is a new bound on the Fourier growth of width-3, oblivious, read-once branching programs. Specifically, we show that for any f : {0,1}^n -> {0,1} computed by such a branching program, and k in [n], sum_{|s|=k} |hat{f}(s)| < n^2 * (O(\log n))^k, where f(x) = sum_s hat{f}(s) (-1)^ is the standard Fourier transform over Z_2^n. The base O(log n) of the Fourier growth is tight up to a factor of log log n.