random
7162 papers tagged with this keyword
Upper tail large deviations of regular subgraph counts in sparse random regular graphs
For a $Δ$-regular connected graph $H$ we obtain the upper tail large deviations speed and rate function for the number of copies of $H$ in $\mathbb{G}_n^d$, a random $d$-regular graph on $n$ vertices, in the regime $np^{Δ/2} \gg (\log n)^{1/(v_H-2)}$ and $p \ll 1$, where $p:=d/n$ and $v_H$ is the number of vertices, and also identify the typical structure of $\mathbb{G}_n^d$ conditioned on this upper tail event. For $Δ\geq 3$, the speed and the rate function are same with those in the Erdős-Rényi setting. In contrast, for $Δ=2$ (i.e.~in case of cycles), due to the degree constraint in $\mathbb{G}_n^d$, although speed remains the same the rate function turns out to be different. Moreover, conditioned on the upper tail event of cycle counts in $\mathbb{G}_n^d$, we show that the random graph contains a disjoint collection of clique and biclique like structures with probability approaching to one, a stark contrast compared to the Erdős-Rényi setting.
In the complementary regime, i.e.~for $1 \ll n^{v_H}p^{e_H} \ll (\log n)^{α^*_H/(α^*_H-1)}$, where $e_H$ is the number of edges in $H$, $α^*_H$ is the fractional independence number of $H$, and $H$ is a strictly balanced graph (and thus including the regular case), we show that the upper tail large deviations speed and rate function for the number of copies of $H$ match with those in the Erdős-Rényi setting. In this regime, we further show that almost all of the unlabeled copies of $H$ conditioned on the upper tail event are edge disjoint.
On Arithmetic-Progression Biases in Random Partitions: Limiting Laws and the Li-Zhao Conjecture
Let $\cR=\{r+jm:j\geq0\}$ and $\cS=\{s+jm:j\geq0\}$ be disjoint arithmetic progressions of positive integers, and choose a partition of $n$ uniformly at random. We determine the limiting probability that the number of parts in $\cR$ is larger than the number of parts in $\cS$. More precisely, if $a=r/m$ and $b=s/m$, then this probability converges to $I_{1/2}(a,b)$, the regularized incomplete beta function. We prove the stronger distributional statement that, after multiplication by the partition saddle-point parameter, the difference of the two numbers of parts converges to $m^{-1}(\log G_b-\log G_a)$, where $G_a$ and $G_b$ are independent gamma variables. The proof uses the Boltzmann conditioning device for random partitions, a $q$-gamma evaluation of the unconditioned characteristic function, and a uniform Fourier estimate showing that conditioning on the total size does not change the limit. We also evaluate the finite-truncation limit studied by Li and Zhao. The two evaluations coincide, proving their conjecture on the commutation of the truncation and large-size limits.
The achromatic number of Paley graphs, Peisert graphs, and related Cayley graphs
The achromatic number of a graph $G$, denoted $χ_a (G)$, is the maximum number of colors in a proper vertex coloring of $G$. Generalizing a result of Maistrelli and Penman, we prove that for sufficiently large $q$, the achromatic number of the $k$-th power Paley graph is at least $\frac{q-1}{k} + 1$. This lower bound gives the correct order of magnitude in $q$, but determining the correct dependence on $k$ remains open for all $k \geq 3$. The achromatic number of Peisert graphs is shown to be $\frac{q+1}{2}$. These bounds are consequences of a general theorem involving Cayley graphs whose generating set is a union of cyclotomic classes. Using character sums and Weil's bound, we prove an upper bound that improves the standard trivial bound $\binom{ χ_a (G) }{2} \leq e(G)$. Finally, we consider random Cayley-sum graphs on an abelian group $Γ$ of order $n$, where elements are included in the generating set independently with probability $\frac{1}{2}$. We prove that the achromatic and pseudochromatic number of such a graph are both at most $\frac{(\sqrt{3}+o(1))n}{\sqrt{\log_2 n}}$ with high probability.
Almost every Boolean function has central threshold degree
The polynomial threshold degree (or threshold degree) of a Boolean function is the least degree of a real polynomial whose sign represents it. We prove that almost every Boolean function has polynomial threshold degree among the central one or two integers. More precisely, a uniformly chosen Boolean function on $n$ variables has threshold degree either $\lfloor n/2\rfloor$ or $\lceil n/2\rceil$, except with probability at most $\exp(-c2^{n/2}/n)$, for an absolute constant $c>0$ and all sufficiently large $n$. The paper proves a 1991 conjecture of Aspnes, Beigel, Furst and Rudich. It also proves a 1991 conjecture of Wang and Williams for even $n$ and its main concentration assertion for odd $n$. A parity example gives a lower bound on the exceptional probability whose exponent differs from ours by a factor of order $n$. The proof relates polynomial sign representation to recovery of Reed--Muller codes from random erasures, and combines sharp-threshold estimates with bounds on generalized Hamming weights and minimum distance.
Erdős-Sós via random cyclic orderings
We present a simplified proof of the Erdős-Sós conjecture on trees in graphs, which is based on its recent resolution by GPT-6 Astra. At the same time, we also give another proof of the corresponding conjecture for antidirected trees in digraphs, posed by Addario-Berry, Havet, Linhares Sales, Reed, and Thomassé, and we simplify recent proofs of Kalai's conjecture for tight trees in hypergraphs. Our proof uses the language of random cyclic orderings in order to take advantage of symmetries that were obscured in the original argument, and we hope that the reader will find it intuitive. We also characterize the extremal examples for Kalai's conjecture, and we generalize it to the setting of so-called fully directed $r$-uniform hypergraphs, whose edges are ordered $r$-tuples.
Locked Polyomino Tilings
A locked $t$-omino tiling is a grid tiling by $t$-ominoes such that, if you remove any pair of tiles, the only way to fill in the remaining $2t$ grid cells with $t$-ominoes is to use the same two tiles in the exact same configuration as before. We exclude degenerate cases where there is only one tiling overall due to small dimensions. It is a classic (and straightforward) result that finite grids do not admit locked 2-omino tilings. In this paper, we construct explicit locked $t$-omino tilings for $t\geq3$ on square and rectangular grids of various dimensions. We show that locked 3- and 4-omino tilings exist on finite square grids of arbitrarily large size. The result for 4-omino tilings in particular is remarkable because they are so rare and difficult to construct: Only a single tiling is known to exist on any grid up to size $40\times40$. We also find locked 5- and 6-omino tilings, and provide a computer-assisted proof that locked 7-omino tilings do not exist on any rectangular grid. In a weighted version of the problem where vertices of the grid may have weights from the set $\{1,2\}$ that count toward the total tile size, we demonstrate the existence of locked tilings on arbitrarily large square weighted grids with only 6 tiles. Finally, we construct locked $t$-omino tilings of the infinite grid for arbitrarily large $t$.
Locked $t$-omino tilings arise as obstructions to widely used political redistricting algorithms in a model of redistricting where the underlying census geography is a grid graph. Most prominent is the ReCom Markov chain, which takes a random walk on the space of redistricting plans by iteratively merging and splitting pairs of districts (tiles) at a time. Locked $t$-omino tilings are isolated states in the state space of ReCom. The constructions in this paper are counterexamples to the meta-conjecture that ReCom is irreducible on graphs of practical interest.
Approximating Kemeny's constant via spectral sparsification
Kemeny's constant of a network quantifies the expected time for a random walk to reach a target vertex chosen according to the stationary distribution, independently of the starting vertex. The purpose of this work is twofold. First, we introduce the class of slightly regular networks, whose orthogonal degree vector is an eigenvector of the Laplacian, and derive an exact expression for Kemeny's constant for this class, extending previous results beyond regular networks. Second, we address the efficient approximation of Kemeny's constant using spectral sparsification. We establish multiplicative bounds relating Kemeny's constant of a network to that of any of its spectral sparsifiers, and use these bounds to design an approximation algorithm. While standard exact approaches based on a full spectral decomposition or on the computation of a pseudoinverse have $O(n^3)$ complexity, our method has complexity $O(n^2\log^{1+o(1)}n)$ for fixed approximation parameters, and is therefore nearly quadratic in the number of vertices even for dense networks.
The longest increasing subsequences of Luce permutations and permutons
We investigate the longest increasing subsequences (LIS) of random permutations distributed under the Luce model with weights $θ_1^{(n)} \ge \ldots \ge θ_n^{(n)}$. Our first main result provides a general formula for the growth rate of the LIS. For example, we get that the LIS of Luce permutations with Sukhatme weights $θ_k^{(n)} = n-k+1$ grows like $\sqrt{n}$. We also give simple conditions for the LIS to be of order $n^β$, for any given $β\in[1/2,1)$. Our second main result states that, if the weights are of the form $θ_k^{(n)} = f\big( (k-1)/n \big)$ for a sufficiently regular function $f$, then the LIS divided by $2\sqrt{n}$ converges in probability to a constant $K(f)$ described by a variational problem. Throughout, we make extensive use of coupling techniques; in particular, to prove our second main result, we argue that Luce random permutations behave similarly to random permutations sampled from the ``Luce permuton''.
Facets of the X-ray polytope of doubly stochastic matrices
We study the polytope of anti-diagonal X-rays of k x k doubly stochastic matrices, the image of the Birkhoff polytope under the map that sums the entries along each anti-diagonal. Its facets other than the coordinate facets are the minimal joint tail bounds for lattice laws of a sum of two dependent uniform random variables. We give an exact facet criterion in terms of the allowed edges of a tight graph. For rays of the underlying tropical fan, total support of the tight graph suffices for a facet but is not necessary, as an example with k=19 shows. Up to affine functions and positive scaling, the potentials that define a facet form a polytope whose vertices are its rays. Facets propagate by padding and by grid refinement under explicit hypotheses. We give infinite families of facets from no step potential, facets of quadratic height, superpolynomially many facets, and a classification of three-step potentials. Total support of a step potential is characterized by a Hall-type condition on block starts; with computer assistance, total support decides facets of step potentials for k<=26. The facet counts up to k=12 are found directly for k<=3 and, for k>=4, come from an enumeration by a provably complete algorithm; an independent exact recomputation reproduces its ray lists for 5<=k<=12.
Kelly's conjectures on Condorcet winner probabilities for odd electorates
Let $C(m,n)$ denote the probability that, among $n$ independent uniformly random permutations of $m$ labels, there exists a label that precedes every other label in a strict majority of the permutations. We prove that $C(m,n)>C(m,n+2)$ for $m \ge 3$ and odd $n \ge 1$, and that $C(m,n)>C(m+1,n)$ for $m \ge 2$ and odd $n \ge 3$. These results establish Kelly's two monotonicity conjectures for odd electorates. The proofs use integral representations involving probabilities associated with sums of independent Bernoulli random variables.
Hitting time for loose Hamilton cycles in the random hypergraph process
The random hypergraph process starts with an empty hypergraph on $n$ vertices and adds a uniformly chosen missing edge at each step. For even $n$, we prove that, with high probability, a $3$-uniform loose Hamilton cycle appears when the last isolated vertex disappears. The approach developed here adapts arguments from Kahn's work on perfect matchings. It can also be extended to obtain the corresponding hitting-time theorems for $r=2$ and every fixed $r\ge4$ satisfying $(r-1)\mid n$. Furthermore, it can be used to estimate the numbers of loose Hamilton cycles at these hitting times for every fixed $r\ge2$, up to a factor of $\exp\{o(n)\}$.
An entropy proof of the commonality of apices of trees
A graph $H$ is called common if the number of monochromatic copies of $H$ in a $2$-edge-colouring of a large complete graph is asymptotically minimized by the random colouring with an equal proportion of each colour. Grzesik, Lee, Lidický and Volec conjectured that every $k$-apex of any connected Sidorenko graph is common. Recently, Kr{á}l', Krnc and Lamaison proved that the $k$-apex of any tree is common for $k\leq 5$. In this paper, we extend their result by showing that for every positive integer $k$, the $k$-apex of any tree or any forest with average degree at least one is common. Our proof involves an application of the entropy method.
Balanced subgraph counts and quasirandomness: a proof of the Janson--Sós conjecture
Let $F$ be a finite simple graph with $m$ vertices and at least two edges, and let $p\in(0,1]$. We prove that a graph sequence is $p$-quasirandom if and only if the symmetrized counts of copies of $F$ across every balanced partition into $m$ classes have the asymptotics predicted by the random graph with edge probability $p$. This settles a conjecture of Janson and Sós for all such $F$, including graphs with isolated vertices. The proof completes their graph-limit reduction by a concavity argument for a sum of fractional powers determined by the degree sequence of $F$. We also show that a symmetrized subgraph kernel of positive mean can be additive in its coordinates only when the underlying graphon is constant almost everywhere. This rigidity result yields a quasirandomness criterion for generalized restricted subgraph counts, resolving a further conjecture of Janson.
Resilience of rainbow Hamilton cycles in pseudorandom graphs
We prove that every spanning subgraph $H$ of an $n$-vertex $(p,β)$-bijumbled graph, satisfying $δ(H)\geq(1/2+\varepsilon)pn$, where $\varepsilon > 0$ is arbitrarily small yet fixed, admits a rainbow Hamilton cycle under any globally $μpn$-bounded colouring, provided that $β\leq cpn$ and $pn\geq M$, where $μ$, $c$, and $M$ are positive constants depending solely on $\varepsilon$; in particular, $pn$ may be taken to be a sufficiently large constant. This answers a question of Coulson, Keevash, Perarnau and Yepremyan for random graphs and extends it to deterministic pseudorandom hosts.
In fact, we prove that $H$ admits at least $(apn)^n$ rainbow Hamilton cycles, where $a > 0$ is a constant depending solely on $\varepsilon$. If $pn\geq D\log n$, then, upon fixing a coloured $H$, retaining each edge independently with probability $D\log n/(pn)$ preserves rainbow Hamiltonicity asymptotically almost surely. The logarithmic degree requirement is needed only for this percolation conclusion.
All three conclusions also hold with rainbowness replaced by avoidance of prescribed pairs of edges, provided that each edge has at most $μpn$ conflicting partners. Our proofs use spread measures and the lopsided local lemma.
Inverting Dependently Random Tournaments
Let $\text{inv}(T)$ be the inversion index of the tournament $T$. Answering a question of Behague, we show that $\text{inv}(T_1\to T_2)=\text{inv}(T_1)+\text{inv}(T_2)$ does not even hold asymptotically.
A Matroidal Framework for Random Access in Linear Codes
We study random access in linear codes when codeword symbols are sampled independently and uniformly with replacement. A code is recovery balanced if the expected number of samples required to recover a codeword coordinate is the same for every coordinate. We express the recovery expectations in terms of the associated matroid and derive a coordinatewise identity relating these values for a code and its dual. This proves that a linear code is recovery balanced if and only if its dual is recovery balanced, establishing a conjecture of Gruica et al. We further define the recovery balance for arbitrary matroids and show that asymptotically almost every sparse paving matroid is not recovery balanced.
Forbidden graded posets in middle levels: supersaturation, enumeration and randomness
Let $P$ be a fixed finite poset whose maximal chains all have $h$ elements, and let $\mathcal W_{n,h}$ be the union of $h$ consecutive middle levels of $2^{[n]}$. We prove that the maximum size of a $P$-free subfamily of $\mathcal W_{n,h}$ is $(π(P)+o(1))M_n$ for a constant $π(P)$, where $M_n=\binom n{\lfloor n/2\rfloor}$, and that the number of $P$-free subfamilies is $2^{(π(P)+o(1))M_n}$. For connected $P$ we prove supersaturation above $(π(P)+ε)M_n$ and determine, in terms of the coordinate patterns of the copies of $P$ in the window, a range of $p$ in which the largest $P$-free subfamily of the random family $\mathcal W_{n,h}(p)$ has size $(π(P)+o(1))pM_n$ and a range in which it has size $(h+o(1))pM_n$; the two ranges meet whenever all copies have the same pattern. For the Boolean poset $Q_d$, which has height $d+1$, the threshold order in the corresponding window $\mathcal W_{n,d+1}$ is $n^{-d/(2^d-1)}$, and $π_d=π(Q_d)$ satisfies $(d+1)λ_d\leπ_d\le d+1-\binom d{\lfloor d/2\rfloor}^{-1}$, where $λ_d$ is the vertex Turán density of $d$-dimensional subcubes in the hypercube. Together with a construction of Ellis, Ivan and Leader this gives $d+1-π_d=\exp(-Θ(d))$. For the diamond, known bounds give $2.1479\leπ_2\le2.15121$.
Sprinkling for random regular graphs and Kim-Vu's sandwich conjecture
We show that the uniform random $d$-regular graph model on $n$ vertices is, in total variation distance, asymptotically close to the union of $d$ uniformly random edge-disjoint perfect matchings, provided that $n$ is even and $ω(1) = d \leq n^{1/10}$. This confirms a conjecture by Hollom, Lichev, Mond, Portier, and Wang for this range of parameters. This result has a couple of interesting consequences. First, this shows that random $d$-regular graphs on $n$ vertices exhibit sprinkling properties for some range of parameters, thereby confirming a conjecture by Isaev, McKay, Southwell and Zhukovskii in that range. Second, we give a short alternative proof of the upper side of Kim-Vu's sandwich conjecture in the sparse regime.
Berry--Esseen Bounds for Faces in Unimodular Triangulations of Random Polytopes
Consider the number of faces arising from unimodular triangulations of random polytopes associated with Erdős--Rényi graphs. Focusing on two-dimensional faces of regular unimodular triangulations of random symmetric edge polytopes, we derive precise asymptotic expansions for the expectation and variance and then establish a quantitative central limit theorem with an explicit Berry--Esseen bound. The proof combines an algebraically derived characterization of faces in terms of local graph configurations with a detailed combinatorial--geometric study of the relevant configurations. The normal approximation is established using the discrete Malliavin--Stein method. We also determine the asymptotic orders of the expectation and variance and establish a quantitative central limit theorem for the number of faces in each fixed higher dimension. These results identify a common local mechanism underlying the fluctuations of face numbers and extend the distributional theory of random symmetric edge polytopes [Donzelmann et al., arXiv:2603.10650] beyond edge statistics.
Solid-angle formulas for simplicial cones and random shifted sections of polytopes
We consider simplicial cones $C = \operatorname{pos}(v_1, \ldots, v_n) \subseteq \mathbb{R}^n$ whose generators have a Gram matrix of the form $(\langle v_i, v_j \rangle)_{i,j=1}^n = \operatorname{Id}_n+\sum_{j=1}^p λ_j u_j u_j^T$, where $p\in \{0,\ldots, n\}$, $λ_1, \ldots, λ_p > -1$ and $u_1,\ldots, u_p \in \mathbb{R}^n$ are orthogonal unit vectors. We derive an explicit $p$-dimensional integral formula for the solid angle of $C$ involving the analytic continuation of the standard normal distribution function.
As an application, explicit formulas are obtained for the expected face numbers of sections of acute orthocentric simplices and crosspolytopes with uniformly distributed linear subspaces translated by a fixed vector. The translation vector is assumed to lie outside the affine hulls of all proper faces of the polytope. All solid angles entering these formulas are expressed as integrals of dimension at most two. This application includes regular simplices and regular crosspolytopes as special cases.