# Why are mod_m gates interesting?

Ryan Williams just posted his lower bound on ACC, the class of problems that have constant depth circuits with unbounded fan-in and gates AND, OR, NOT and MOD_m for all possible m's.

What's so special about MOD_m gates?

• They allow one to simulate arithmetic over any ring Z_m.
• Before Ryan's result, throwing MOD_m gates to the mix gave the first class for which the known lower bounds did not work.

Is there any other natural reason to study MOD_m gates?

-

$ACC^0$ is a natural complexity class.

1) Barrington showed that computation over non-solvable monoids capture $NC^1$ while over solvable monoids capture $ACC^0$.

2) Recently, Hansen and Koucky proved a beautiful result that poly-sized constant width planar branching programs are exactly $ACC^0$. Without the planarity condition, we of course get Barrington's result characterizing $NC^1$.

So the difference between $ACC^0$ and $NC^1$ is group-theoretic on one hand and topological on the other.

Added: Dana, a simple example of a solvable group is $S_4$, the symmetric group over elements. Without getting into details, any solvable group has a series whose quotients happen to be cyclic. This cyclic structure gets reflected as mod gates while building a circuit to solve word problems over the group.

On planarity, one would like to believe that planarity may impose restrictions/bottlenecks in the flow of information. This is not always true: for example, variations of planar 3SAT are known to be NP-complete. However, in smaller classes, these restrictions are more "likely" to hold.

In similar vein, Wigderson showed NL/poly=UL/poly using the isolation lemma. We do not know how to derandomize the isolation lemma over arbitrary DAGs to get NL=UL, but we know how to do so for planar DAGs.

-
Thanks a lot for the information! I'd love to hear more about the intuition for these results. As to my question: your argument is basically that $NC^1$ [O(log n) depth, gates AND, OR, NOT] is natural, and $ACC$ is a slight variation of it (to solvable rather than non-solvable monoids, or to planar rather than non-planar branching programs). Could you elaborate a little: give examples of interesting monoids for computation, and how their solvability matters? Is there an a-priori motivation to take interest in whether a branching program is planar or not? –  Dana Moshkovitz Nov 9 '10 at 15:15
To complement: 1) Computation over aperiodic monoids capture $AC^0$ (Barrington and Thérien). 2) Upward planar branching programs capture $AC^0$ (Barrington, Lu, Miltersen, Skyum). –  Kristoffer Arnsfelt Hansen Nov 9 '10 at 15:55
@Vinay: Are you sure that the result NL/poly=UL/poly is due to Wigderson? –  Dai Le Jul 8 '11 at 21:22

Perhaps this isn't really an answer to your question. But to just give one example of why sometimes $\bmod m$ gates (for composite $m$) are more powerful than $\bmod p$ gates:

Consider the class of constant depth circuits that consist only of $\bmod p$ gates, and inputs and constants at the leaves. Then, one can easily show that the OR function (for example) cannot be computed by such circuits, regardless of the size of the circuit. (This is because any such circuit computes a low degree polynomial over $\mathbb{F}_p$, and the degree of OR is $n$).

However, if we consider circuits that consist only of $\bmod m$ gates where $m$ has at least two distinct prime factors, there is a depth $2$ circuit (of exponential size) for the OR function.

And prior to Ryan's result, $AC^0[\bmod 6]$ was I guess the smallest class for which we didn't have any decent lower bounds.

-
Addendum to the last sentence: It was already known that computing $MOD_q$ with constant-depth circuits using AND, OR, NOT, and $MOD_p$ gates for primes $p \ne q$ required an exponential number of gates. (There's an extension to relatively prime composites as well.) Since 6 is the smallest composite of two distinct primes, $MOD_6$ is the "easiest"-to-compute function that no exponential lower bound was known for. –  Daniel Apon Nov 10 '10 at 16:10