next up previous contents
Next: Sensitive Functions Up: Boolean Functions Previous: Nondeterministic Decision Tree Complexity   Contents


Nondeterministically Evasive Functions

With an understanding of decision tree complexity we can now prove a lower bound on a class of evasive functions. In analogy to evasive functions whose deterministic decision tree complexity is $N$ we call functions with nondeterministic decision tree complexity $N$ Nondeterministically evasive. Every nondeterministically evasive functions is evasive.

Theorem 4.3.1   $\Omega(\sqrt{N})$ oracle queries are required to compute any nondeterministically evasive $N$-bit Boolean function in the bounded error setting.


\begin{proof}
% latex2html id marker 2304We will prove that any nondeterminist...
...e queries are
required to compute $f$\ in the bounded error setting.
\end{proof}

This proof establishes that nondeterministically evasive functions have inputs that are sensitive to negation on at least half of their bits; the result then follows from Lemma [*]. Not all evasive functions are nondeterministically evasive. OR is an example of a nondeterministically evasive function; since Beals et al. provide an $O(\sqrt{N})$ algorithm to compute the OR of $N$ bits, this lower bound is asymptotically tight.



Matthew Hayward 2008-04-26