CS Paper Radar

On Top-Down and Local Lower Bounds for $\mathrm{AC^0}$ Circuits

Gülce Kardeş, Benjamin Rossman

cs.CC ·submitted 2026-09-01 18:25 UTC · v1
theory

Abstract

Classical lower bounds for $\mathrm{AC^0}$ circuits proceed bottom-up by simplifying or approximating gates beginning at the input layer. We introduce a complementary top-down model called the Chopping Game, played by adversaries Spoiler and Duplicator on the sets of $0$- and $1$-inputs of a Boolean function. In each round, Spoiler keeps at least a $1/m$-fraction of one side, and Duplicator arbitrarily restricts the other; Spoiler seeks to minimize (and Duplicator to maximize) the number of rounds until some coordinate separates the two remaining sets. Every depth-$d$, fan-in-$m$ circuit induces a $d$-round winning strategy for Spoiler, while Duplicator strategies that survive $d$ rounds formalize top-down lower-bound arguments. Through the Chopping Game and using the polynomial-approximation method, we first obtain the classical lower bound for depth-$d$ $\mathrm{AC^0}$ circuits in a top-down fashion. We then consider a $k$-local variant of the Chopping Game, which relaxes Spoiler's win condition by requiring a separating coordinate within each Hamming ball of radius $k$, rather than a single coordinate globally. We put forward a conjecture that the $d$-round $k$-local Chopping Game for $\mathrm{PARITY}$ requires $m = n^{ω(1)}$ in the regime $d \ll k \ll n$. We prove such a lower bound $m \ge n^{Ω(k^{1/d}/d)}$ when Spoiler is restricted to so-called affine strategies, a class of strategies that achieves the best known upper bounds. Finally, we formulate a version of the $k$-local Chopping Game on $n$-regular graphs of girth $>2k$, and we conjecture a graph-theoretic analogue of ``$\mathrm{PARITY} \notin \mathrm{AC^0}$''.

Comment
22 pages

Closest papers

Nearest neighbours by meaning across the whole index.

Optimal Shallow Circuits for Majority

cs.CC ·2026-09-27 ·Victor Lecomte, Prasanna Ramakrishnan
theoryhighly cited similarity 0.69
Abstract

Four decades on, Håstad's classical $2^{Ω(n^{1/(d-1)})}$ lower bound for depth-$d$ circuits computing Parity remains the best known $\mathrm{AC}^0$ circuit lower bound for any explicit function. Majority has long been a compelling candidate for stronger lower bounds: the most natural circuits computing it are substantially larger than those for Parity and have repeatedly been conjectured to be optimal. We present a simple construction, found by GPT-6 Astra, of depth-$d$ circuits of size $2^{O(n^{1/(d-1)})}$ for any symmetric function. This result settles the asymptotic $\mathrm{AC}^0$ circuit complexity of Majority, matching Håstad's lower bound. The proof draws inspiration from well-loved combinatorial tools, including the color-coding technique of Alon, Yuster, and Zwick (1995).

0.33

Sumset Structure in Local Computation

cs.CC ·2026-09-19 ·Alexander Golovnev, Mohit Gurumukhani
publishedtheory similarity 0.65
Abstract

We introduce and study a new model of decision trees that lies at the frontier of provable circuit lower bounds. We study $\ell$-local decision trees, in which each internal node queries an $\ell$-local function of the input. We show that sufficiently strong lower bounds for $\ell$-local decision trees would imply several breakthrough circuit lower bounds, including super-linear size lower bounds for log-depth circuits and improved bounds for unrestricted-depth circuits. Previously, such consequences were known to follow from strong lower bounds for depth-$3$ circuits, which has been the main prior approach in attempting to prove these results. Since local decision trees are strictly weaker than depth-$3$ circuits, this provides a formally easier route to the same circuit lower-bound consequences. We prove essentially optimal lower bounds for a weaker variant that we call oblivious $\ell$-local decision trees, where all nodes at the same depth query the same function. Our lower bounds follow from a new technique that uncovers sumset structure in local maps, and our hard functions are sumset dispersers, sumset condensers, and directional affine dispersers. Along the way, we give an explicit construction of a sumset condenser with small entropy loss. As an additional contribution, we initiate the study of degree-$2$ decision trees, in which each query is a quadratic polynomial of the input. We show that proving lower bounds in this model is a natural stepping stone toward constructing dispersers for degree-$2$ variety sources, which imply improved circuit lower bounds for unrestricted depth circuits. We prove nearly maximal lower bounds for the oblivious variant of the model.

0.35

Short Point Lists for CNFs and Stronger Constant-Depth Circuit Hardness

cs.CC math.CO ·2026-10-19 ·Xin Li, Yan Zhong
publishedtheory similarity 0.65
Abstract

We give a short-description decomposition of bounded-width CNFs. For every fixed $\varepsilon>0$, every width-$b$ CNF on $m$ variables is an exact disjoint union of at most $2^{\varepsilon m}$ width-$b$ CNFs, each determined by $2^{O_\varepsilon(b)}\log(m+1)$ satisfying assignments. For every fixed $d\ge3$, we prove a worst-case size lower bound of $2^{Ω_d((n\log n)^{1/(d-1)})}$ against depth-at-most-$d$ unbounded-fan-in AND/OR circuits with input negations for three polynomial-time Boolean families: inverse trace over binary fields, the Goldreich--Tal trilinear function, and OpenAI's polynomial evaluator with its original parameters. All parameter bits count as inputs. For inverse trace, every such circuit of size at most $S$, where $n\le S\le2^{c_dn^{1/(d-1)}}$, has uniform sign correlation at most $2^{-η_dn\log\log S/(\log S)^{d-1}}$. Here $c_d,η_d>0$ depend only on $d$. This improves on the classical parity correlation bound. The lower bounds combine the decomposition with restriction arguments; the correlation bound additionally uses common partial decision trees. We also construct exact nonuniform AND/OR circuits for inverse trace of size $2^{O(n^{2/3})}$ at depth eight and $2^{O(\sqrt n\log n)}$ at depth nine. For every fixed integer $h\ge2$, the construction gives depth-$(4h+1)$ circuits of size $2^{O_h(n^{1/h}\log n)}$. The upper bounds implement finite-field evaluation and interpolation using circuits that enumerate intermediate states.

0.52

A Top-Down Depth-Four Circuit Lower Bound for Majority

cs.CC ·2026-08-10 ·Hao Wu, Yaqiao Li
theory similarity 0.63
Abstract

We present a top-down depth-four circuit lower bound for Majority function by extending recent work of Göös, Riazanov, Sofronova, and Sokolov (FOCS 2023), who gave a top-down proof of a depth-four circuit lower bound for Parity which relies on the robust sunflower to construct a mirror set and the block unpredictability to find the local limits. The main challenge for the case of Majority is to construct a corresponding mirror set, the difference is that to flip the value of Majority function, one may have to flip many bits of the input Boolean string, while for Parity, flipping one bit suffices. We avoid this flipping by considering slices of the Boolean cube, that is, Boolean strings of fixed Hamming weight approximately $n/2$.

0.32

Top-Down Lower Bounds for All Depths

cs.CC ·2026-09-30 ·Oliver Korten
publishedtheory similarity 0.62
Abstract · 20 pages

We prove that Parity requires $2^{n^{Ω(1)}}$ size De Morgan circuits of constant depth using a new method which is completely "top-down" in the sense of [HJP95]. The proof relies crucially on the core ideas developed in a line of work [HJP95, PPZ99, MW19, GRSS24] which previously established top-down lower bounds for circuits of depth 3 and 4. We first present a proof of a lower bound $\exp(n^{3^{-d}})$. In this case, nearly all of the relevant combinatorial ideas necessary for the proof are already present in some form in [GRSS24]. We then present two extensions of this argument, the first achieving a lower bound $\exp(ε_d n^{1/(2d-2)})$ for some $ε_d>0$ depending only on $d$, and the second achieving the essentially tight lower bound $\exp(ε_d n^{1/(d-1)})$. These improved results each hinge on establishing a key lemma which quantifies the extent to which a high entropy random variable in $\{0,1\}^n$ will look close to uniform after projecting it onto a random small set of coordinates $R \subseteq [n]$.

0.46

Group Isomorphism and the Polylogarithmic-Time Hierarchy: Depth-2$\frac{1}{2}$ Circuits and Lower Bounds

cs.CC cs.DSmath.GR ·2026-08-26 ·Joshua A. Grochow, Gülce Kardeş, Michael Levet
publishedtheory similarity 0.61
Abstract · New result: lower-bound against CNF-of-XORs. 69 pages

In this paper, we investigate the low-depth circuit complexity of Group Isomorphism in the multiplication (Cayley) table model. We prove the first circuit lower bounds for Group Isomorphism: namely, we show that every family of depth-$2$ Boolean circuits deciding Group Isomorphism requires quasipolynomial-size. (We also extend this to a limited class of depth-3 circuits, namely $\textsf{CNF}$-of-$\textsf{XOR}$s.) We complement this with upper bounds of uniform depth-$2\frac{1}{2}$ circuits of quasipolynomial-size: depth 3 circuits with poly-logarithmic bottom fan-in. A sequence of previous results from 1970-2025 progressively reduced the circuit depth from polynomial to $3\frac{1}{2}$; all of these results relied on the generator-enumerator strategy and, in fact, applied more generally to quasigroups. In contrast, our depth-$2\frac{1}{2}$ construction follows a fundamentally different strategy that exploits structure more specific to groups. We guess a composition series for each group, together with generators for its terms and the isomorphism types of its composition factors. We then inductively verify that the corresponding extensions at each level of the two composition series are compatible. A central part in this approach brings to bear the extensive work on the Short Presentation Conjecture, in tandem with the algorithmic theory of group extensions and cohomology.

0.42

Depth Hierarchies for Symmetric Algebraic Circuits

cs.CC ·2026-11-17 ·Anuj Dawar, Prateek Dwivedi, Benedikt Pago, Tim Seppelt
similarity 0.59
Abstract · submitted to STOC 2027

Algebraic complexity studies the cost of computing families of polynomials by algebraic circuits, with the central goal of proving super-polynomial circuit lower bounds and separations between different models of algebraic computation. Since such lower bounds remain elusive for general circuits, a common strategy is to focus on restricted models. A recent line of work (ICALP 2020, ITCS & STOC 2026) introduces and studies symmetry as a syntactic circuit restriction, showing exponential symmetric lower bounds for the permanent, as well as separations between symmetric analogues of algebraic formulas ($\mathsf{VF}$), skew circuits ($\mathsf{VS}$), and circuits ($\mathsf{VP}$). We investigate the role of depth as a complexity parameter in symmetric circuits. For every fixed $Δ\geq 1$, we exhibit a family of polynomials computable by symmetric circuits of polynomial size and depth $2Δ+3$, but requiring size $n^{Ω(\log n)}$ at depth $Δ$. We also define symmetric versions of the bounded-depth algebraic complexity classes $\mathsf{VAC}^k$, for all $k \in \mathbb{N}$, that we call $\mathsf{symVAC}^k$, and establish two strict hierarchies: Firstly, a hierarchy within $\mathsf{symVAC}^0$, and secondly, a hierarchy of the form $\mathsf{symVAC}^k \subsetneq \mathsf{symVAC}^{k+1}$ for all $k$. The latter shows that symmetric circuits of depth $(\log n)^{k+1}$ are strictly more powerful than depth-$(\log n)^k$ circuits. Moreover, we place the bounded-depth classes within the landscape of symmetric algebraic complexity developed by Dwivedi, Pago and Seppelt (STOC 2026). We complement this picture by introducing $\mathsf{symVNP}$, a symmetric analogue of $\mathsf{VNP}$, and establishing $\mathsf{symVNP}$-completeness of the permanent under symmetric projections.

0.29

A Witnessing Duality for ReLU Network Depth

cs.CC cs.LGcs.LO ·2026-10-28 ·Dimitrios Myrisiotis
theory similarity 0.58
Abstract · 132 pages, 1 figure, 3 tables

A depth lower bound for ReLU networks asserts that every sufficiently shallow network fails on some input. What does a proof of such a bound compute? We study this question for exact computation of the maximum of $n$ real inputs, with depth measured by the number of hidden layers. For every separately fixed prime $q$ and depth bound $K$, we give a uniform $\mathsf{TC}^0$ refuter for rational networks with at most $K$ hidden layers and more than $q^K$ inputs, provided that the reduced denominators of their nonbias weights are not divisible by $q$. Biases may be arbitrary rationals. The refuter finds a rational input on which the network fails, and its correctness is provable in $\mathbf{VTC}^0$, the bounded-arithmetic theory of uniform constant-depth threshold computation. The underlying numerical depth lower bound follows from Averkov, Hojny and Merkert (ICLR 2025), extending Haase, Hertrich and Loho (ICLR 2023); our contribution is an internal proof that makes its geometric argument into a feasible refuter. On the construction side, $\mathbf{VTC}^0$ verifies locally checkable certificates for the balanced integer-weight maximum network of depth $\lceil\log_2 n\rceil$. At $q=2$, the construction and refuter meet: for each fixed $K$, the theory verifies that integer-nonbias-weight networks with arbitrary rational biases can compute the maximum with at most $K$ hidden layers exactly when $n\le 2^K$. We also give certified rational constructions of polynomial encoding length and depth $O(\log n/\log\log n)$. These results instantiate a witnessing duality: bounded proofs of certified existence and of refutation correspond, respectively, to threshold-computable constructors and refuters whose guarantees the theory proves.

0.26