English

Tree tribes and lower bounds for switching lemmas

Computational Complexity 2017-03-02 v1

Abstract

We show tight upper and lower bounds for switching lemmas obtained by the action of random pp-restrictions on boolean functions that can be expressed as decision trees in which every vertex is at a distance of at most tt from some leaf, also called tt-clipped decision trees. More specifically, we show the following: \bullet If a boolean function ff can be expressed as a tt-clipped decision tree, then under the action of a random pp-restriction ρ\rho, the probability that the smallest depth decision tree for fρf|_{\rho} has depth greater than dd is upper bounded by (4p2t)d(4p2^{t})^{d}. \bullet For every tt, there exists a function gtg_{t} that can be expressed as a tt-clipped decision tree, such that under the action of a random pp-restriction ρ\rho, the probability that the smallest depth decision tree for gtρg_{t}|_{\rho} has depth greater than dd is lower bounded by (c0p2t)d(c_{0}p2^{t})^{d}, for 0pcp2t0\leq p\leq c_{p}2^{-t} and 0dcdlogn2tlogt0\leq d\leq c_{d}\frac{\log n}{2^{t}\log t}, where c0,cp,cdc_{0},c_{p},c_{d} are universal constants.

Cite

@article{arxiv.1703.00043,
  title  = {Tree tribes and lower bounds for switching lemmas},
  author = {Jenish C. Mehta},
  journal= {arXiv preprint arXiv:1703.00043},
  year   = {2017}
}
R2 v1 2026-06-22T18:31:25.733Z