当前位置: X-MOL 学术Comput. Complex. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
The Average Sensitivity of Bounded-Depth Formulas
computational complexity ( IF 1.4 ) Pub Date : 2017-07-05 , DOI: 10.1007/s00037-017-0156-0
Benjamin Rossman

AbstractWe show that unbounded fan-in Boolean formulas of depth d + 1 and size s have average sensitivity $${O(\frac{1}{d} \log s)^d}$$O(1dlogs)d. In particular, this gives a tight $${2^{\Omega(d(n^{1/d}-1))}}$$2Ω(d(n1/d-1)) lower bound on the size of depth d + 1 formulas computing the parity function. These results strengthen the corresponding $${2^{\Omega(n^{1/d})}}$$2Ω(n1/d) and $${O(\log s)^d}$$O(logs)d bounds for circuits due to Håstad (Proceedings of the 18th annual ACM symposium on theory of computing, ACM, New York, 1986) and Boppana (Inf Process Lett 63(5): 257–261, 1997). Our proof technique studies a random process where the switching lemma is applied to formulas in an efficient manner.

中文翻译:

有界深度公式的平均灵敏度

摘要我们表明深度 d + 1 和大小 s 的无界扇入布尔公式具有平均灵敏度 $${O(\frac{1}{d} \log s)^d}$$O(1dlogs)d。特别是,这给出了紧缩的 $${2^{\Omega(d(n^{1/d}-1))}}$$2Ω(d(n1/d-1)) 下界计算奇偶校验函数的深度 d + 1 公式。这些结果加强了相应的 $${2^{\Omega(n^{1/d})}}$$2Ω(n1/d) 和 $${O(\log s)^d}$$O(logs由于 Håstad(第 18 届年度 ACM 计算理论研讨会论文集,ACM,纽约,1986 年)和 Boppana(Inf Process Lett 63(5):257–261,1997 年),电路的 )d 界限。我们的证明技术研究了一个随机过程,其中以有效的方式将切换引理应用于公式。
更新日期:2017-07-05
down
wechat
bug