arXiv subjects

Vishesh Jain

Publications and source records attributed to Vishesh Jain.

66 records · Page 4Linked to original sources

arXiv2019

On the counting problem in inverse Littlewood--Offord theory

Let $ε_1, \dotsc, ε_n$ be i.i.d. Rademacher random variables taking values $\pm 1$ with probability $1/2$ each. Given an integer vector $\boldsymbol{a} = (a_1, \dotsc, a_n)$, its concentration probability is the quantity $ρ(\boldsymbol{a}):=\sup_{x\in \mathbb{Z}}\Pr(ε_1 a_1+\dots+ε_n a_n = x)$. The Littlewood-Offord problem asks for bounds on $ρ(\boldsymbol{a})$ under various hypotheses on $\boldsymbol{a}$, whereas the inverse Littlewood-Offord problem, posed by Tao and Vu, asks for a characterization of all vectors $\boldsymbol{a}$ for which $ρ(\boldsymbol{a})$ is large. In this paper, we study the associated counting problem: How many integer vectors $\boldsymbol{a}$ belonging to a specified set have large $ρ(\boldsymbol{a})$? The motivation for our study is that in typical applications, the inverse Littlewood-Offord theorems are only used to obtain such counting estimates. Using a more direct approach, we obtain significantly better bounds for this problem than those obtained using the inverse Littlewood--Offord theorems of Tao and Vu and of Nguyen and Vu. Moreover, we develop a framework for deriving upper bounds on the probability of singularity of random discrete matrices that utilizes our counting result. To illustrate the methods, we present the first `exponential-type' (i.e., $\exp(-n^c)$ for some positive constant $c$) upper bounds on the singularity probability for the following two models: (i) adjacency matrices of dense signed random regular digraphs, for which the previous best known bound is $O(n^{-1/4})$ due to Cook; and (ii) dense row-regular $\{0,1\}$-matrices, for which the previous best known bound is $O_{C}(n^{-C})$ for any constant $C>0$ due to Nguyen.

math.CO↗

arXiv2019

Uniformity-independent minimum degree conditions for perfect matchings in hypergraphs

In this note, we prove that there exists a universal constant $c=\frac{43}{50}$ such that for every $k\in \mathbb{N}$ and every $d<k/2$, every $k$-uniform hypergraph on $n$ vertices and with minimum $d$-degree at least $(c+o_n(1))\binom{n-d}{k-d}$ contains a perfect matching. This is the first such bound which is independent of $k$, and therefore, improves all previously known bounds when $k$ is large. Our approach is based on combining the seminal work of Alon et al. with known bounds on a conjectured probabilistic inequality due to Feige.

math.CO↗

arXiv2018

Towards the linear arboricity conjecture

The linear arboricity of a graph $G$, denoted by $\text{la}(G)$, is the minimum number of edge-disjoint linear forests (i.e. forests in which every connected component is a path) in $G$ whose union covers all the edges of $G$. A famous conjecture due to Akiyama, Exoo, and Harary from 1980 asserts that $\text{la}(G)\leq \lceil (Δ(G)+1)/2 \rceil$, where $Δ(G)$ denotes the maximum degree of $G$. This conjectured upper bound would be best possible, as is easily seen by taking $G$ to be a regular graph. In this paper, we show that for every graph $G$, $\text{la}(G)\leq \fracΔ{2}+O(Δ^{2/3-α})$ for some $α> 0$, thereby improving the previously best known bound due to Alon and Spencer from 1992. For graphs which are sufficiently good spectral expanders, we give even better bounds. Our proofs of these results further give probabilistic polynomial time algorithms for finding such decompositions into linear forests.

math.CO↗

arXiv2018

Mean-field approximation, convex hierarchies, and the optimality of correlation rounding: a unified perspective

The free energy is a key quantity of interest in Ising models, but unfortunately, computing it in general is computationally intractable. Two popular (variational) approximation schemes for estimating the free energy of general Ising models (in particular, even in regimes where correlation decay does not hold) are: (i) the mean-field approximation with roots in statistical physics, which estimates the free energy from below, and (ii) hierarchies of convex relaxations with roots in theoretical computer science, which estimate the free energy from above. We show, surprisingly, that the tight regime for both methods to compute the free energy to leading order is identical. More precisely, we show that the mean-field approximation is within $O((n\|J\|_{F})^{2/3})$ of the free energy, where $\|J\|_F$ denotes the Frobenius norm of the interaction matrix of the Ising model. This simultaneously subsumes both the breakthrough work of Basak and Mukherjee, who showed the tight result that the mean-field approximation is within $o(n)$ whenever $\|J\|_{F} = o(\sqrt{n})$, as well as the work of Jain, Koehler, and Mossel, who gave the previously best known non-asymptotic bound of $O((n\|J\|_{F})^{2/3}\log^{1/3}(n\|J\|_{F}))$. We give a simple, algorithmic proof of this result using a convex relaxation proposed by Risteski based on the Sherali-Adams hierarchy, automatically giving sub-exponential time approximation schemes for the free energy in this entire regime. Our algorithmic result is tight under Gap-ETH. We furthermore combine our techniques with spin glass theory to prove (in a strong sense) the optimality of correlation rounding, refuting a recent conjecture of Allen, O'Donnell, and Zhou. Finally, we give the tight generalization of all of these results to $k$-MRFs, capturing as a special case previous work on approximating MAX-$k$-CSP.

cs.LG↗

arXiv2018

On the number of Hadamard matrices via anti-concentration

Many problems in combinatorial linear algebra require upper bounds on the number of solutions to an underdetermined system of linear equations $Ax = b$, where the coordinates of the vector $x$ are restricted to take values in some small subset (e.g. $\{\pm 1\}$) of the underlying field. The classical ways of bounding this quantity are to use either a rank bound observation due to Odlyzko or a vector anti-concentration inequality due to Halász. The former gives a stronger conclusion except when the number of equations is significantly smaller than the number of variables; even in such situations, the hypotheses of Halász's inequality are quite hard to verify in practice. In this paper, using a novel approach to the anti-concentration problem for vector sums, we obtain new Halász-type inequalities which beat the Odlyzko bound even in settings where the number of equations is comparable to the number of variables. In addition to being stronger, our inequalities have hypotheses which are considerably easier to verify. We present two applications of our inequalities to combinatorial (random) matrix theory: (i) we obtain the first non-trivial upper bound on the number of $n\times n$ Hadamard matrices, and (ii) we improve a recent bound of Deneanu and Vu on the probability of normality of a random $\{\pm 1\}$ matrix.

math.PR↗

arXiv2018

On the k-planar local crossing number

Given a fixed positive integer $k$, the $k$-planar local crossing number of a graph $G$, denoted by $\text{LCR}_k(G)$, is the minimum positive integer $L$ such that $G$ can be decomposed into $k$ subgraphs, each of which can be drawn in a plane such that no edge is crossed more than $L$ times. In this note, we show that under certain natural restrictions, the ratio $\text{LCR}_k(G)/\text{LCR}_1(G)$ is of order $1/k^2$, which is analogous to a recent result of Pach et al. for the $k$-planar crossing number (defined as the minimum positive integer $C$ for which there is a $k$-planar drawing of $G$ with $C$ total edge crossings). As a corollary of our proof we show that, under similar restrictions, one may obtain a $k$-planar drawing of $G$ with \emph{both} the total number of edge crossings as well as the maximum number of times any edge is crossed essentially matching the best known bounds. Our proof relies on the crossing number inequality and several probabilistic tools such as concentration of measure and the Lovász local lemma.

math.CO↗

arXiv2018

1-factorizations of pseudorandom graphs

A $1$-factorization of a graph $G$ is a collection of edge-disjoint perfect matchings whose union is $E(G)$. A trivial necessary condition for $G$ to admit a $1$-factorization is that $|V(G)|$ is even and $G$ is regular; the converse is easily seen to be false. In this paper, we consider the problem of finding $1$-factorizations of regular, pseudorandom graphs. Specifically, we prove that an $(n,d,λ)$-graph $G$ (that is, a $d$-regular graph on $n$ vertices whose second largest eigenvalue in absolute value is at most $λ$) admits a $1$-factorization provided that $n$ is even, $C_0\leq d\leq n-1$ (where $C_0$ is a universal constant), and $λ\leq d^{1-o(1)}$. In particular, since (as is well known) a typical random $d$-regular graph $G_{n,d}$ is such a graph, we obtain the existence of a $1$-factorization in a typical $G_{n,d}$ for all $C_0\leq d\leq n-1$, thereby extending to all possible values of $d$ results obtained by Janson, and independently by Molloy, Robalewska, Robinson, and Wormald for fixed $d$. Moreover, we also obtain a lower bound for the number of distinct $1$-factorizations of such graphs $G$ which is off by a factor of $2$ in the base of the exponent from the known upper bound. This lower bound is better by a factor of $2^{nd/2}$ than the previously best known lower bounds, even in the simplest case where $G$ is the complete graph. Our proofs are probabilistic and can be easily turned into polynomial time (randomized) algorithms.

math.CO↗

arXiv2018

The Vertex Sample Complexity of Free Energy is Polynomial

We study the following question: given a massive Markov random field on $n$ nodes, can a small sample from it provide a rough approximation to the free energy $\mathcal{F}_n = \log{Z_n}$? Results in graph limit literature by Borgs, Chayes, Lovász, Sós, and Vesztergombi show that for Ising models on $n$ nodes and interactions of strength $Θ(1/n)$, an $ε$ approximation to $\log Z_n / n$ can be achieved by sampling a randomly induced model on $2^{O(1/ε^2)}$ nodes. We show that the sampling complexity of this problem is {\em polynomial in} $1/ε$. We further show a polynomial dependence on $ε$ cannot be avoided. Our results are very general as they apply to higher order Markov random fields. For Markov random fields of order $r$, we obtain an algorithm that achieves $ε$ approximation using a number of samples polynomial in $r$ and $1/ε$ and running time that is $2^{O(1/ε^2)}$ up to polynomial factors in $r$ and $ε$. For ferromagnetic Ising models, the running time is polynomial in $1/ε$. Our results are intimately connected to recent research on the regularity lemma and property testing, where the interest is in finding which properties can tested within $ε$ error in time polynomial in $1/ε$. In particular, our proofs build on results from a recent work by Alon, de la Vega, Kannan and Karpinski, who also introduced the notion of polynomial vertex sample complexity. Another critical ingredient of the proof is an effective bound by the authors of the paper relating the variational free energy and the free energy.

cs.LG↗

arXiv2018

Approximating Partition Functions in Constant Time

We study approximations of the partition function of dense graphical models. Partition functions of graphical models play a fundamental role is statistical physics, in statistics and in machine learning. Two of the main methods for approximating the partition function are Markov Chain Monte Carlo and Variational Methods. An impressive body of work in mathematics, physics and theoretical computer science provides conditions under which Markov Chain Monte Carlo methods converge in polynomial time. These methods often lead to polynomial time approximation algorithms for the partition function in cases where the underlying model exhibits correlation decay. There are very few theoretical guarantees for the performance of variational methods. One exception is recent results by Risteski (2016) who considered dense graphical models and showed that using variational methods, it is possible to find an $O(εn)$ additive approximation to the log partition function in time $n^{O(1/ε^2)}$ even in a regime where correlation decay does not hold. We show that under essentially the same conditions, an $O(εn)$ additive approximation of the log partition function can be found in constant time, independent of $n$. In particular, our results cover dense Ising and Potts models as well as dense graphical models with $k$-wise interaction. They also apply for low threshold rank models.

cs.LG↗

arXiv2018

The Mean-Field Approximation: Information Inequalities, Algorithms, and Complexity

The mean field approximation to the Ising model is a canonical variational tool that is used for analysis and inference in Ising models. We provide a simple and optimal bound for the KL error of the mean field approximation for Ising models on general graphs, and extend it to higher order Markov random fields. Our bound improves on previous bounds obtained in work in the graph limit literature by Borgs, Chayes, Lovász, Sós, and Vesztergombi and another recent work by Basak and Mukherjee. Our bound is tight up to lower order terms. Building on the methods used to prove the bound, along with techniques from combinatorics and optimization, we study the algorithmic problem of estimating the (variational) free energy for Ising models and general Markov random fields. For a graph $G$ on $n$ vertices and interaction matrix $J$ with Frobenius norm $\| J \|_F$, we provide algorithms that approximate the free energy within an additive error of $εn \|J\|_F$ in time $\exp(poly(1/ε))$. We also show that approximation within $(n \|J\|_F)^{1-δ}$ is NP-hard for every $δ> 0$. Finally, we provide more efficient approximation algorithms, which find the optimal mean field approximation, for ferromagnetic Ising models and for Ising models satisfying Dobrushin's condition.

cs.LG↗

arXiv2017

A Counterexample to the "Majority is Least Stable" Conjecture

We exhibit a linear threshold function in 5 variables with strictly smaller noise stability (for small values of the correlation parameter) than the majority function on 5 variables, thereby providing a counterexample to the "Majority is Least Stable" Conjecture of Benjamini, Kalai, and Schramm.

cs.CC↗

arXiv2014

On discontinuity of planar optimal transport maps

Consider two bounded domains $Ω$ and $Λ$ in $\mathbb{R}^{2}$, and two sufficiently regular probability measures $μ$ and $ν$ supported on them. By Brenier's theorem, there exists a unique transportation map $T$ satisfying $T_\#μ=ν$ and minimizing the quadratic cost $\int_{\mathbb{R}^{n}}|T(x)-x|^{2}dμ(x)$. Furthermore, by Caffarelli's regularity theory for the real Monge--Ampère equations, if $Λ$ is convex, $T$ is continuous. We study the reverse problem, namely, when is $T$ discontinuous if $Λ$ fails to be convex? We prove a result guaranteeing the discontinuity of $T$ in terms of the geometries of $Λ$ and $Ω$ in the two-dimensional case. The main idea is to use tools of convex analysis and the extrinsic geometry of $\partialΛ$ to distinguish between Brenier and Alexandrov weak solutions of the Monge--Ampère equation. We also use this approach to give a new proof of a result due to Wolfson and Urbas. We conclude by revisiting an example of Caffarelli, giving a detailed study of a discontinuous map between two explicit domains, and determining precisely where the discontinuities occur.

math.AP↗