Bounded Depth Arithmetic Circuits: Counting and Closure
David M Barrington1999, Lecture Notes in Computer Science
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
Constant-depth arithmetic circuits have been defined and studied in [AAD97, ABL98]; these circuits yield the function classes #AC 0 and GapAC 0. These function classes in turn provide new characterizations of the computational power of threshold circuits, and provide a link between the circuit classes AC 0 (where many lower bounds are known) and TC 0 (where essentially no lower bounds are known). In this paper, we resolve several questions regarding the closure properties of #AC 0 and GapAC 0. Counting classes are usually characterized in terms of problems of counting paths in a class of graphs (simple paths in directed or undirected graphs for #P, simple paths in directed acyclic graphs for #L, or paths in bounded-width graphs for GapNC 1). It was shown in [BLMS98] that complete problems for depth k Boolean AC 0 can be obtained by considering the reachability problem for width k grid graphs. It would be tempting to conjecture that #AC 0 could be characterized by counting paths in bounded-width grid graphs. We disprove this, but nonetheless succeed in characterizing #AC 0 by counting paths in another family of bounded-width graphs.
Related papers
On the depth complexity of the counting functions
Information Processing Letters, 1990
We use Karchmer and Wigderson's recent characterization of circuit depth in terms of communication complexity to design shallow Boolean circuits for the counting functions. We show that the MOD, counting function on n arguments can be computed by Boolean networks which contain negations and binary OR-and AND-gates in depth c logrn, where c A 2.881. This is an improvement over the obvious depth upper bound of 3 logan. We can also design circuits for the MOD, and MOD,, functions having depth 3.475 logan and 4.930 logan, respectively.
On the computational power of depth 2 circuits with threshold and modulo gates
Proceedings of the twenty-sixth annual ACM …, 1994
We investigate the computational power of depth-2 circuits consisting of MOD' gates at the bottom and a threshold gate with arbitrary weights at the top (for short, threshold-MOD' circuits) and circuits with two levels of MOD gates (MODp-MOD4 circuits). In particular, we will show the following results: (i) For all prime numbers p and integers q,r, it holds that if p divides r but not q then all threshold-MOD4 circuits for MOD' have exponentially many nodes. (ii) For all integers r, all problems computable by depth-2 {AND,OR,NOT} circuits of polynomial size have threshold-MOD' circuits with polynomially many edges. (iii) There is a problem computable by depth 3 {AND, OR, NOT} circuits of linear size and constant bottom fan-in which for all r needs threshold-MOD' circuits with exponentially many nodes.
Threshold Circuits of Bounded Depth
Journal of Computer and System Sciences, 1993
We examine a powerful model of parallel computation: polynomial size threshold circuits of bounded depth (the gates compute threshold functions with polynomial weights). Lower bounds are given to separate polynomial size threshold circuits of depth 2 from polynomial size threshold circuits of depth 3 and from probabilistic polynomial size circuits of depth 2. With regard to the unreliability of bounded depth circuits, it is shown that the class of functions computed reliably with bounded depth circuits of unreliable A, v , 1 gates is narrow. On the other hand, functions computable by bounded depth, polynomial-size threshold circuits can also be computed by such circuits of unreliable threshold gates. Furthermore we examine to what extent imprecise threshold gates (which behave unpredictably near the threshold value) can compute nontrivial functions in bounded depth and a bound is given for the permissible amount of imprecision. We also discuss threshold quantifiers and prove an undefinability result for graph connectivity.
Limiting negations in bounded-depth circuits: An extension of Markov's theorem
Information Processing Letters, 2004
From a theorem of Markov, the minimum number of negation gates in a circuit sufficient to compute any collection of Boolean functions on n variable is ℓ = ⌈log(n + 1)⌉. Santha and Wilson [SIAM Journal of Computing 22 : 294-302 (1993)] showed that in some classes of bounded-depth circuits ℓ negation gates are no longer sufficient for some explicitly defined Boolean function. In this paper, we consider a general class of bounded-depth circuits in which each gate computes an arbitrary monotone Boolean function or its negation. Our purpose is to extend the theorem of Markov for such a general class of circuits. We first show that a lower bound shown by Santha and Wilson becomes an extension of Markov's lower bound by a small refinement. Then, we present tight upper bounds on the number of negations for computing an arbitrary collection of Boolean functions.
On the Depth Complexity of the Counting Functions, 35 Info. Processing Letters 325 (1990)
We use Karchmer and Wigderson's recent characterization of circuit depth in terms of communication complexity to design shallow Boolean circuits for the counting functions.
A size-depth trade-off for the analog computation of Boolean functions
Information Processing Letters, 1996
Satisfiability and Derandomization for Small Polynomial Threshold Circuits
Electron. Colloquium Comput. Complex., 2018
A polynomial threshold function (PTF) is defined as the sign of a polynomial p : {0, 1} → R. A PTF circuit is a Boolean circuit whose gates are PTFs. We study the problems of exact and (promise) approximate counting for PTF circuits of constant depth. Satisfiability (#SAT). We give the first zero-error randomized algorithm faster than exhaustive search that counts the number of satisfying assignments of a given constant-depth circuit with a super-linear number of wires whose gates are s-sparse PTFs, for s almost quadratic in the input size of the circuit; here a PTF is called s-sparse if its underlying polynomial has at most s monomials. More specifically, we show that, for any large enough constant c, given a depth-d circuit with (n2−1/c)-sparse PTF gates that has at most n1+εd wires, where εd depends only on c and d, the number of satisfying assignments of the circuit can be computed in randomized time 2n−nε with zero error. This generalizes the result by Chen, Santhanam and Srini...
Communication complexity towards lower bounds on circuit depth
2002
Karchmer, Raz, and Wigderson, 1991, discuss the circuit depth complexity of n bit Boolean functions constructed by composing up to d = log n= log log n levels of k = log n bit boolean functions. Any such function is in AC 1. They conjecture that circuit depth is additive under composition, which would imply that any (bounded fan-in) circuit for this problem requires dk 2 (log 2 n= log log n) depth. This would separate AC 1 from NC 1. They recommend using the communication game characterization of circuit depth. In order to develop techniques for using communication complexity to prove circuit lower bounds, they suggest an intermediate communication complexity problem which they call the Universal Composition Relation. We give an almost optimal lower bound of dk O(d 2 (k log k) 1=2) for this problem. In addition, we present a proof, directly in terms of communication complexity, that there is a function on k bits requiring (k) circuit depth. Although this fact can be easily established using a counting argument, we hope that the ideas in our proof will be incorporated more easily into subsequent arguments which use communication complexity to prove circuit depth bounds.
Computational Complexity of Some Enumeration Problems About Uniformly Sparse Boolean Network Automata
2006
We study the computational complexity of counting the fixed point configurations (FPs), the predecessor configurations and the ancestor configurations in certain classes of network automata viewed as discrete dynamical systems. Some early results of this investigation are presented in [38, 39]. In particular, it is proven in [39] that both exact and approximate counting of FPs in the two closely related classes of Boolean network automata, called Sequential and Synchronous Dynamical Systems (SDSs and SyDSs, respectively), are computationally intractable problems when each node is required to update according to a monotone Boolean function. In the present paper, we further strengthen those results by showing that the intractability of exact enumeration of FPs of a monotone Boolean SDS or SyDS still holds even when (i) the monotone update rules are restricted to linear threshold functions, and (ii) the underlying graph is uniformly sparse. By uniform sparseness we mean that every node in the graph has its degree bounded by Ç´½µ for a small value of constant. In particular, we prove that exactly enumerating FPs in such SDSs and SyDSs remains #P-complete even when no node degree exceeds ¿. Among other consequences, we show that this result also implies intractability of determining the exact memory capacity of discrete Hopfield networks with uniformly sparse and nonnegative integer weight matrices.
On Threshold Circuits and Polynomial Computation
SIAM Journal on Computing, 1992
A Threshold Circuit consists of an acyclic digraph of unbounded fanin, where each node computes a threshold function or its negation. This paper investigates the computational power of Threshold Circuits. A surprising relationship is uncovered between Threshold Circuits and another class of unbounded fanin circuits which are denoted Finite Field Z P (n) Circuits, where each node computes either multiple sums or products of integers modulo a prime P (n). In particular, it is proved that all functions computed by Threshold Circuits of size S(n) ≥ n and depth D(n) can also be computed by Z P (n) Circuits of size O(S(n) log S(n) + nP (n) log P (n)) and depth O(D(n)). Furthermore, it is shown that all functions computed by Z P (n) Circuits of size S(n) and depth D(n) can be computed by Threshold Circuits of size O(1 2 (S(n) log P (n)) 1+) and depth O(1 5 D(n)). These are the main results of this paper. There are many useful and quite surprising consequences of this result. For example, integer reciprocal can be computed in size n O(1) and depth O(1). More generally, any analytic function with a convergent rational polynomial power series (such as sine, cosine, exponentiation, square root, and logarithm) can be computed within accuracy 2 −n c , for any constant c, by Threshold Circuits of polynomial size and constant depth. In addition, integer and polynomial division, FFT, polynomial interpolation, Chinese Remaindering, all the elementary symmetric functions, banded matrix inverse, and triangular Toeplitz matrix inverse can be exactly computed by Threshold Circuits of polynomial size and constant depth. All these results and simulations hold for polytime uniform circuits. This paper also gives a corresponding simulation of logspace uniform Z P (n) Circuits by logspace uniform Threshold Circuits requiring an additional multiplying factor of O(log log log P (n)) depth. Finally, purely algebraic methods for lower bounds for Z P (n) Circuits are developed. Using degree arguments, a Depth Hierarchy Theorem for Z P (n) Circuits is proved: for any S(n) ≥ n, D(n) = O(S(n) c) for some constant c < 1, and prime P (n) where 6(S(n)/D(n)) D(n) < P (n) ≤ 2 n , there exists explicitly constructible functions computable by Z P (n) Circuits of size S(n) and depth D(n), but provably not computable by Z P (n) Circuits of size S(n) c and depth o(D(n)) for any constant c ≥ 1.
Related papers
An exponential lower bound for depth 3 arithmetic circuits
1998
We prove the rst exponential lower bound on the size of any depth 3 arithmetic circuit with unbounded fanin computing an explicit function (the determinant) over an arbitrary nite eld. This answers an open problem of N91] and NW95] for the case of nite elds. We intepret here arithmetic circuits in the algebra of polynomials over the given eld. The proof method involves a new argument on the rank of linear functions, and a group symmetry on polynomials vanishing at certain nonsingular matrices, and could be of independent interest.
Depth reduction for circuits with a single layer of modular counting gates
2009
We consider the class of constant depth AND/OR circuits augmented with a layer of modular counting gates at the bottom layer, ie AC 0° MOD m\ bf AC^ 0 ∘\ bf MOD _m circuits. We show that the following holds for several types of gates GG: by adding a gate of type GG at the output, it is possible to obtain an equivalent probabilistic depth 2 circuit of quasipolynomial size consisting of a gate of type GG at the output and a layer of modular counting gates, ie G° MOD m G ∘\ bf MOD _m circuits.
Descriptive complexity of circuit-based counting classes
2021
In this thesis, we study the descriptive complexity of counting classes based on Boolean circuits. In descriptive complexity, the complexity of problems is studied in terms of logics required to describe them. The focus of research in this area is on identifying logics that can express exactly the problems in specific complexity classes. For example, problems are definable in ESO, existential second-order logic, if and only if they are in NP, the class of problems decidable in nondeterministic polynomial time. In the computation model of Boolean circuits, individual circuits have a fixed number of inputs. Circuit families are used to allow for an arbitrary number of input bits. A priori, the circuits in a family are not uniformly described, but one can impose this as an additional condition, e.g., requiring that there is an algorithm constructing them. For any circuit there is a function counting witnesses (or proofs) for the circuit producing the output 1. Consequently, any class o...
Energy Complexity and Depth of Threshold Circuits
Lecture Notes in Computer Science, 2009
In the paper we show that there is a close relationship between the energy complexity and the depth of threshold circuits computing any Boolean function although they have completely different physical meanings. Suppose that a Boolean function f can be computed by a threshold circuit C of energy complexity e and hence at most e threshold gates in C output "1" for any input to C. We then prove that the function f can be computed also by a threshold circuit C of depth 2e + 1 and hence the parallel computation time of C is 2e + 1. If the size of C is s, that is, there are s threshold gates in C, then the size s of C is s = 2es + 1. Thus, if the size s of C is polynomial in the number n of input variables, then the size s of C is polynomial in n, too.
New algorithms and lower bounds for circuits with linear threshold gates
Proceedings of the forty-sixth annual ACM symposium on Theory of computing, 2014
Let ACC• THR be the class of constant-depth circuits comprised of AND, OR, and MODm gates (for some constant m > 1), with a bottom layer of gates computing arbitrary linear threshold functions. This class of circuits can be seen as a "midpoint" between ACC (where we know nontrivial lower bounds) and depth-two linear threshold circuits (where nontrivial lower bounds remain open). We give an algorithm for evaluating an arbitrary symmetric function of 2 n o(1) ACC • THR circuits of size 2 n o(1) , on all possible inputs, in 2 n • poly(n) time. Several consequences are derived: • The number of satisfying assignments to an ACC • THR circuit of subexponential size can be computed in 2 n−n ε time (where ε > 0 depends on the depth and modulus of the circuit). • NEXP does not have quasi-polynomial size ACC • THR circuits, and NEXP does not have quasipolynomial size ACC• SYM circuits. Nontrivial size lower bounds were not known even for AND• OR • THR circuits. • Every 0-1 integer linear program with n Boolean variables and s linear constraints is solvable in 2 n−Ω(n/((logM)(log s) 5)) • poly(s, n, M) time with high probability, where M upper bounds the bit complexity of the coefficients. (For example, 0-1 integer programs with weights in [−2 poly(n) , 2 poly(n) ] and poly(n) constraints can be solved in 2 n−Ω(n/ log 6 n) time.) Impagliazzo, Paturi, and Schneider [IPS13] recently gave an algorithm forÕ(n) constraints; ours is the first asymptotic improvement over exhaustive search for for up to subexponentially many constraints. We also present an algorithm for evaluating depth-two linear threshold circuits (a.k.a., THR • THR) with exponential weights and 2 n/24 size on all 2 n input assignments, running in 2 n • poly(n) time. This is evidence that non-uniform lower bounds for THR • THR are within reach.
Energy and depth of threshold circuits
Theoretical Computer Science, 2010
In the paper we show that there is a close relationship between the energy complexity and the depth of threshold circuits computing any Boolean function although they have completely different physical meanings. Suppose that a Boolean function f can be computed by a threshold circuit C of energy complexity e and hence at most e threshold gates in C output "1" for any input to C. We then prove that the function f can be computed also by a threshold circuit C of depth 2e + 1 and hence the parallel computation time of C is 2e + 1. If the size of C is s, that is, there are s threshold gates in C, then the size s of C is s = 2es + 1. Thus, if the size s of C is polynomial in the number n of input variables, then the size s of C is polynomial in n, too.
Exponential lower bounds on the size of constant-depth threshold circuits with small energy complexity
Theoretical Computer Science, 2008
A complexity measure for threshold circuits, called the energy complexity, has been proposed to measure an amount of energy consumed during computation in the brain. Biological neurons need more energy to transmit a ''spike'' than not to transmit one, and hence the energy complexity of a threshold circuit is defined as the number of gates in the circuit that output ''1'' during computation. Since the firing activity of neurons in the brain is quite sparse, the following question arises: what Boolean functions can or cannot be computed by threshold circuits with small energy complexity. In the paper, we partially answer the question, that is, we show that there exists a trade-off among three complexity measures of threshold circuits: the energy complexity, size, and depth. The trade-off implies an exponential lower bound on the size of constant-depth threshold circuits with small energy complexity for a large class of Boolean functions.
Demystifying the border of depth-3 algebraic circuits
2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS), 2022
Border complexity of polynomials plays an integral role in GCT (Geometric complexity theory) approach to P = NP. It tries to formalize the notion of 'approximating a polynomial' via limits (Bürgisser FOCS'01). This raises the open question VP ? = VP; as the approximation involves exponential precision which may not be efficiently simulable. Recently (Kumar ToCT'20) proved the universal power of the border of top-fanin-2 depth-3 circuits (Σ [2] ΠΣ). Here we answer some of the related open questions. We show that the border of bounded top-fanin depth-3 circuits (Σ [k] ΠΣ for constant k) is relatively easy-it can be computed by a polynomial size algebraic branching program (ABP). There were hardly any de-bordering results known for prominent models before our result. Moreover, we give the first quasipolynomial-time blackbox identity test for the same. Prior best was in PSPACE (Forbes,Shpilka STOC'18). Also, with more technical work, we extend our results to depth-4. Our de-bordering paradigm is a multi-step process; in short we call it DiDIL-divide, derive, induct, with limit. It 'almost' reduces Σ [k] ΠΣ to special cases of read-once oblivious algebraic branching programs (ROABPs) in any-order.
An exponential lower bound for homogeneous depth four arithmetic circuits with bounded bottom fanin
have recently shown that an exponential lower bound for depth four homogeneous circuits with bottom layer of × gates having sublinear fanin translates to an exponential lower bound for a general arithmetic circuit computing the permanent. Motivated by this, we examine the complexity of computing the permanent and determinant via homogeneous depth four circuits with bounded bottom fanin. We show here that any homogeneous depth four arithmetic circuit with bounded bottom fanin computing the permanent (or the determinant) must be of exponential size.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.