Separation of AC$^0[\oplus]$ Formulas and Circuits

Abstract: This paper gives the first separation between the power of {\em formulas} and {\em circuits} of equal depth in the $\mathrm{AC}<sup>0[\oplus]$ basis (unbounded fan-in AND, OR, NOT and MOD$_2$ gates). We show, for all $d(n) \le O(\frac{\log n}{\log\log n})$, that there exist {\em polynomial-size depth-$d$ circuits} that are not equivalent to {\em depth-$d$ formulas of size $n<sup>{o(d)}$} (moreover, this is optimal in that $n<sup>{o(d)}$ cannot be improved to $n<sup>{O(d)}$). This result is obtained by a combination of new lower and upper bounds for {\em Approximate Majorities}, the class of Boolean functions ${0,1}ⁿ \to {0,1}$ that agree with the Majority function on $3/4$ fraction of inputs. $\mathrm{AC}<sup>0[\oplus]$ formula lower bound: We show that every depth-$d$ $\mathrm{AC}<sup>0[\oplus]$ formula of size $s$ has a {\em $1/8$-error polynomial approximation} over $\mathbb{F}_2$ of degree $O(\frac{1}{d}\log s)<sup>{d-1}$. This strengthens a classic $O(\log s)<sup>{d-1}$ degree approximation for \underline{circuits} due to Razborov. Since the Majority function has approximate degree $\Theta(\sqrt n)$, this result implies an $\exp(\Omega(dn<sup>{1/2(d-1)}))$ lower bound on the depth-$d$ $\mathrm{AC}<sup>0[\oplus]$ formula size of all Approximate Majority functions for all $d(n) \le O(\log n)$. Monotone $\mathrm{AC}<sup>0$ circuit upper bound: For all $d(n) \le O(\frac{\log n}{\log\log n})$, we give a randomized construction of depth-$d$ monotone $\mathrm{AC}<sup>0$ circuits (without NOT or MOD$_2$ gates) of size $\exp(O(n<sup>{1/2(d-1)}))$ that compute an Approximate Majority function. This strengthens a construction of \underline{formulas} of size $\exp(O(dn<sup>{1/2(d-1)}))$ due to Amano.