Santhoshini Velusamy

Chi-Ning Chou, Alexander Golovnev, Madhu Sudan et al.

We consider the approximability of constraint satisfaction problems in the streaming setting. For every constraint satisfaction problem (CSP) on $n$ variables taking values in $\{0,\ldots,q-1\}$, we prove that improving over the trivial approximability by a factor of $q$ requires $Ω(n)$ space even on instances with $O(n)$ constraints. We also identify a broad subclass of problems for which any improvement over the trivial approximability requires $Ω(n)$ space. The key technical core is an optimal, $q^{-(k-1)}$-inapproximability for the Max $k$-LIN-$\bmod\; q$ problem, which is the Max CSP problem where every constraint is given by a system of $k-1$ linear equations $\bmod\; q$ over $k$ variables. Our work builds on and extends the breakthrough work of Kapralov and Krachun (Proc. STOC 2019) who showed a linear lower bound on any non-trivial approximation of the MaxCut problem in graphs. MaxCut corresponds roughly to the case of Max $k$-LIN-$\bmod\; q$ with ${k=q=2}$. For general CSPs in the streaming setting, prior results only yielded $Ω(\sqrt{n})$ space bounds. In particular no linear space lower bound was known for an approximation factor less than $1/2$ for any CSP. Extending the work of Kapralov and Krachun to Max $k$-LIN-$\bmod\; q$ to $k>2$ and $q>2$ (while getting optimal hardness results) is the main technical contribution of this work. Each one of these extensions provides non-trivial technical challenges that we overcome in this work.

Noah G. Singer, Madhur Tulsiani, Santhoshini Velusamy

For an arbitrary family of predicates $\mathcal{F} \subseteq \{0,1\}^{[q]^k}$ and any $ε> 0$, we prove a single-pass, linear-space streaming lower bound against the gap promise problem of distinguishing instances of Max-CSP$({\mathcal{F}})$ with at most $β+ε$ fraction of satisfiable constraints from instances of with at least $γ-ε$ fraction of satisfiable constraints, whenever Max-CSP$({\mathcal{F}})$ admits a $(γ,β)$-integrality gap instance for the basic LP. This subsumes the linear-space lower bound of Chou, Golovnev, Sudan, Velingker, and Velusamy (STOC 2022), which applies only to a special subclass of CSPs with linear-algebraic structure. (Their result itself generalizes work of Kapralov and Krachun (STOC 2019) for Max-CUT.) Our approach identifies the right ``analytic'' analogues of previously-used linear-algebraic conditions; this yields substantial simplifications while capturing a much larger class of problems. Our lower bound is essentially optimal for single-pass streaming, since: (1) All CSPs admit $(1-ε)$-approximations in quasilinear space, and (2) sublinear-space streaming algorithms can simulate the LP (on bounded-degree instances), giving approximation algorithms when integrality gap instances do not exist. The starting point for our lower bound is a reduction from a "distributional implicit hidden partition'' problem defined by Fei, Minzer, and Wang (STOC 2026) in the context of multi-pass streaming. Our result is an analogue of theirs in the single-pass setting, where we obtain a much stronger (and tight) space lower bound.

Santhoshini Velusamy

The Max-DICUT problem has gained a lot of attention in the streaming setting in recent years, and has so far served as a canonical problem for designing algorithms for general constraint satisfaction problems (CSPs) in this setting. A seminal result of Kapralov and Krachun [STOC 2019] shows that it is impossible to beat $1/2$-approximation for Max-DICUT in sublinear space in the single-pass streaming setting, even on bounded-degree graphs. In a recent work, Saxena, Singer, Sudan, and Velusamy [SODA 2025] prove that the above lower bound is tight by giving a single-pass algorithm for bounded-degree graphs that achieves $(1/2-ε)$-approximation in sublinear space, for every constant $ε>0$. For arbitrary graphs of unbounded degree, they give an $O(1/ε)$-pass $O(\log n)$ space algorithm. Their work left open the question of obtaining $1/2$-approximation for arbitrary graphs in the single-pass setting in sublinear space. We make progress towards this question and give a two-pass algorithm that achieves $(1/2-ε)$-approximation in sublinear space, for every constant $ε>0$.

Amatya Sharma, Santhoshini Velusamy

Non-redundancy, introduced by Bessiere, Carbonnel, and Katsirelos (AAAI 2020), is a structural parameter for Constraint Satisfaction Problems ($\mathsf{CSPs}$) that governs kernelization, exact and approximate sparsification, and exact streaming complexity. It is the largest size of a $\mathsf{CSP}$ instance admitting no smaller subinstance with the same satisfying assignments. We study non-redundancy $\mathsf{NRD}_n(R)$ for Boolean symmetric $\mathsf{CSPs}$ defined by an $r$-ary relation $R$ whose value depends only on Hamming weight. An instance of $\mathsf{CSP}(R)$ has $n$ variables and constraints given by $r$-tuples; a constraint is satisfied exactly when the induced tuple lies in $R$. This class includes natural predicates such as cuts and $k$-SAT clauses. Our main result is a near-complete classification of the asymptotic growth of $\mathsf{NRD}_n(R)$ for symmetric Boolean predicates of arity at most $5$. Using computational experiments and algebraic upper- and lower-bound criteria, we resolve every predicate of arity at most $4$ and all but two predicates of arity $5$. For upper bounds, we introduce $t$-balancedness, a lifted, higher-degree version of the balancedness notion of Chen, Jansen, and Pieterse (Algorithmica 2020). We prove that $t$-balancedness is equivalent to the existence of degree-$t$ multilinear polynomials capturing $R$, and hence implies $\mathsf{NRD}_n(R)=O(n^t)$. For lower bounds, we use Carbonnel's (CP 2022) framework: predicates admitting a special reduction from $k$-ary OR inherit OR's lower bound $Ω(n^k)$. The only unresolved arity-$5$ predicates in our framework have bounds $Ω(n^2)$ and $O(n^3)$; we reduce their exact classification to natural extremal set-system questions.

Amatya Sharma, Santhoshini Velusamy

We study the single-pass streaming complexity of deciding satisfiability of Constraint Satisfaction Problems (CSPs). A CSP is specified by a constraint language $Γ$, that is, a finite set of $k$-ary relations over the domain $[q] = \{0, \dots, q-1\}$. An instance of $\mathsf{CSP}(Γ)$ consists of $m$ constraints over $n$ variables $x_1, \ldots, x_n$ taking values in $[q]$. Each constraint $C_i$ is of the form $\{R_i,(x_{i_1} + λ_{i_1}, \ldots, x_{i_k} + λ_{i_k})\}$, where $R_i \in Γ$ and $λ_{i_1}, \ldots, λ_{i_k} \in [q]$ are constants; it is satisfied if and only if $(x_{i_1} + λ_{i_1}, \ldots, x_{i_k} + λ_{i_k}) \in R_i$, where addition is modulo $q$. In the streaming model, constraints arrive one by one, and the goal is to determine, using minimum memory, whether there exists an assignment satisfying all constraints. For $k$-SAT, Vu (TCS 2024) proves an optimal $Ω(n^k)$ space lower bound, while for general CSPs, Chou, Golovnev, Sudan, and Velusamy (JACM 2024) establish an $Ω(n)$ lower bound; a complete characterization has remained open. We close this gap by showing that the single-pass streaming space complexity of $\mathsf{CSP}(Γ)$ is precisely governed by its non-redundancy, a structural parameter introduced by Bessiere, Carbonnel, and Katsirelos (AAAI 2020). The non-redundancy $\mathsf{NRD}_n(Γ)$ is the maximum number of constraints over $n$ variables such that every constraint $C$ is non-redundant, i.e., there exists an assignment satisfying all constraints except $C$. We prove that the single-pass streaming complexity of $\mathsf{CSP}(Γ)$ is characterized, up to a logarithmic factor, by $\mathsf{NRD}_n(Γ)$.