Searcharxiv⌕ Search

arXiv subjects

Dmitry Ostrovsky

Publications and source records attributed to Dmitry Ostrovsky.

14 recordsLinked to original sources
arXiv2020

A Family of Probability Distributions Consistent with the DOZZ Formula: Towards a Conjecture for the Law of 2D GMC

A three parameter family of probability distributions is constructed such that its Mellin transform is defined over the same domain as the 2D GMC on the Riemann sphere with three insertion points $(\alpha_1,\alpha_2,\alpha_3)$ and satisfies the DOZZ formula in the sense of Kupiainen (Ann. Math. 191 (2020) 81 -- 166). The probability distributions in the family are defined as products of independent Fyodorov-Bouchaud and powers of Barnes beta distributions of types $(2, 1)$ and $(2, 2).$ In the special case of $\alpha_1+\alpha_2+\alpha_3=2Q$ the constructed probability distribution is shown to be consistent with the known small deviation asymptotic of the 2D GMC laws with everywhere positive curvature.

math.PR↗
arXiv2018

A Review of Conjectured Laws of Total Mass of Bacry-Muzy GMC Measures on the Interval and Circle and Their Applications

Selberg and Morris integral probability distributions are long conjectured to be distributions of the total mass of the Bacry-Muzy Gaussian Multiplicative Chaos measures with non-random logarithmic potentials on the unit interval and circle, respectively. The construction and properties of these distributions are reviewed from three perspectives: analytic based on several representations of the Mellin transform, asymptotic based on low intermittency expansions, and probabilistic based on the theory of Barnes beta probability distributions. In particular, positive and negative integer moments, infinite factorizations and involution invariance of the Mellin transform, analytic and probabilistic proofs of infinite divisibility of the logarithm, factorizations into products of Barnes beta distributions, and Stieltjes moment problems of these distributions are presented in detail. Applications are given in the form of conjectured mod-Gaussian limit theorems, laws of derivative martingales, distribution of extrema of $1/f$ noises, and calculations of inverse participation ratios in the Fyodorov-Bouchaud model.

math.PR↗
arXiv2016

A Theory of Intermittency Differentiation of 1D Infinitely Divisible Multiplicative Chaos Measures

A theory of intermittency differentiation is developed for a general class of 1D Infinitely Divisible Multiplicative Chaos measures. The intermittency invariance of the underlying infinitely divisible field is established and utilized to derive a Feynman-Kac equation for the distribution of the total mass of the limit measure by considering a stochastic flow in intermittency. The resulting equation prescribes the rule of intermittency differentiation for a general functional of the total mass and determines the distribution of the total mass and its dependence structure to the first order in intermittency. A class of non-local functionals of the limit measure extending the total mass is introduced and shown to be invariant under intermittency differentiation making the computation of the full high temperature expansion of the total mass distribution possible in principle. For application, positive integer moments and covariance structure of the total mass are considered in detail.

math.PR↗
arXiv2016

Subcritical multiplicative chaos for regularized counting statistics from random matrix theory

For an $N \times N$ random unitary matrix $U_N$, we consider the random field defined by counting the number of eigenvalues of $U_N$ in a mesoscopic arc of the unit circle, regularized at an $N$-dependent scale $\epsilon_N>0$. We prove that the renormalized exponential of this field converges as $N \to \infty$ to a Gaussian multiplicative chaos measure in the whole subcritical phase. In addition, we show that the moments of the total mass converge to a Selberg-like integral and by taking a further limit as the size of the arc diverges, we establish part of the conjectures in \cite{Ost16}. By an analogous construction, we prove that the multiplicative chaos measure coming from the sine process has the same distribution, which strongly suggests that this limiting object should be universal. The proofs are based on the asymptotic analysis of certain Toeplitz or Fredholm determinants using the Borodin-Okounkov formula or a Riemann-Hilbert problem for integrable operators. Our approach to the $L^{1}$-phase is based on a generalization of the construction in Berestycki \cite{Berestycki15} to random fields which are only \textit{asymptotically} Gaussian. In particular, our method could have applications to other random fields coming from either random matrix theory or a different context.

math.PR↗
arXiv2016

A Theory of Intermittency Renormalization of Gaussian Multiplicative Chaos Measures

A theory of intermittency differentiation is developed for a general class of Gaussian Multiplicative Chaos measures including the measure of Bacry and Muzy on the interval and circle as special cases. An exact, non-local functional equation is derived for the derivative of a general functional of the total mass of the measure with respect to intermittency. The formal solution is given in the form of an intermittency expansion and proved to be a renormalized expansion in the centered moments of the total mass of the measure. The full intermittency expansion of the Mellin transform of the total mass is computed in terms of the corresponding expansion of log-moments. The theory is shown to extend to the dependence structure of the measure. For application, the intermittency expansion of the Bacry-Muzy measure on the circle is computed exactly, and the Morris integral probability distribution is shown to reproduce the moments of the total mass and the intermittency expansion, resulting in the conjecture that it is the distribution of the total mass. It is conjectured in general that the intermittency expansion captures the distribution of the total mass uniquely.

math.PR↗
arXiv2016

A Note on Moments of Limit Log Infinitely Divisible Stochastic Measures of Bacry and Muzy

A multiple integral representation of single and joint moments of the total mass of the limit log-infinitely divisible stochastic measure of Bacry and Muzy [$\textit{Comm. Math. Phys.}$ ${\bf 236}$: 449-475, 2003] is derived. The covariance structure of the total mass of the measure is shown to be logarithmic. A generalization of the Selberg integral corresponding to single moments of the limit measure is proposed and shown to satisfy a recurrence relation. The joint moments of the limit lognormal measure, classical Selberg integral with $\lambda_1=\lambda_2=0,$ and Morris integral are represented in the form of multiple binomial sums. For application, low moments of the limit log-Poisson measure are computed exactly and low joint moments of the limit lognormal measure are considered in detail.

math.PR↗
arXiv2016

Structure-Blind Signal Recovery

We consider the problem of recovering a signal observed in Gaussian noise. If the set of signals is convex and compact, and can be specified beforehand, one can use classical linear estimators that achieve a risk within a constant factor of the minimax risk. However, when the set is unspecified, designing an estimator that is blind to the hidden structure of the signal remains a challenging problem. We propose a new family of estimators to recover signals observed in Gaussian noise. Instead of specifying the set where the signal lives, we assume the existence of a well-performing linear estimator. Proposed estimators enjoy exact oracle inequalities and can be efficiently computed through convex optimization. We present several numerical illustrations that show the potential of the approach.

math.ST↗
arXiv2016

On Barnes Beta Distributions and Applications to the Maximum Distribution of the 2D Gaussian Free Field

A new family of Barnes beta distributions on $(0, \infty)$ is introduced and its infinite divisibility, moment determinacy, scaling, and factorization properties are established. The Morris integral probability distribution is constructed from Barnes beta distributions of types $(1,0)$ and $(2,2),$ and its moment determinacy and involution invariance properties are established. For application, the maximum distributions of the 2D gaussian free field on the unit interval and circle with a non-random logarithmic potential are conjecturally related to the critical Selberg and Morris integral probability distributions, respectively, and expressed in terms of sums of Barnes beta distributions of types $(1,0)$ and $(2,2).$

math.PR↗
arXiv2015

On Riemann zeroes, Lognormal Multiplicative Chaos, and Selberg Integral

Rescaled Mellin-type transforms of the exponential functional of the Bourgade-Kuan-Rodgers statistic of Riemann zeroes are conjecturally related to the distribution of the total mass of the limit lognormal stochastic measure of Mandelbrot-Bacry-Muzy. The conjecture implies that a non-trivial, log-infinitely divisible probability distribution is associated with Riemann zeroes. For application, integral moments, covariance structure, multiscaling spectrum, and asymptotics associated with the exponential functional are computed in closed form using the known meromorphic extension of the Selberg integral.

math.PR↗
arXiv2014

Theory of Barnes Beta Distributions

A new family of probability distributions $β_{M, N},$ $M=0\cdots N,$ $N\in\mathbb{N}$ on the unit interval $(0, 1]$ is defined by the Mellin transform. The Mellin transform of $β_{M, N}$ is characterized in terms of products of ratios of Barnes multiple gamma functions, shown to satisfy a functional equation, and a Shintani-type infinite product factorization. The distribution $\logβ_{M, N}$ is infinitely divisible. If $M<N,$ $-\logβ_{M, N}$ is compound Poisson, if $M=N,$ $\logβ_{M, N}$ is absolutely continuous. The integral moments of $β_{M, N}$ are expressed as Selberg-type products of multiple gamma functions. The asymptotic behavior of the Mellin transform is derived and used to prove an inequality involving multiple gamma functions and establish positivity of a class of alternating power series. For application, the Selberg integral is interpreted probabilistically as a transformation of $β_{1, 1}$ into a product of $β^{-1}_{2, 2}s.$

math.PR↗
arXiv2014

On Barnes Beta Distributions, Selberg Integral and Riemann Xi

The theory of Barnes beta probability distributions is advanced and related to the Riemann xi function. The scaling invariance, multiplication formula, and Shintani factorization of Barnes multiple gamma functions are reviewed using the approach of Ruijsenaars and shown to imply novel properties of Barnes beta distributions. The applications are given to the meromorphic extension of the Selberg integral as a function of its dimension and the scaling invariance of the underlying probability distribution. This probability distribution in the critical case is described and conjectured to be the distribution of the derivative martingale. The Jacobi triple product is interpreted probabilistically resulting in an approximation of the Riemann xi function by the Mellin transform of the logarithm of a limit of Barnes beta distributions.

math.PR↗
arXiv2004

Instanton-induced Azimuthal Spin Asymmetry in Deep Inelastic Scattering

It is by now well understood that spin asymmetry in deep inelastic scattering (DIS) can appear if two things are both present: (i) a chirality flip of the struck quark; (ii) a nonzero T-odd phase due to its final state interaction. So far (i) was attributed to a new structure/wave function of the nucleon and (ii) to some gluon exchanges. We propose a new mechanism utilizing strong vacuum fluctuations of the gluon field described semiclasically by instantons, and show that both (i) and (ii) are present. The magnitude of the effect is estimated using known parameters of the instanton ensemble in the QCD vacuum and known structure and fragmentation functions, without any new free parameters. The result agrees in sign and (roughly) in magnitude with the available data on single particle inclusive DIS. Furthermore, our predictions uniquely relate effects for longitudinally and transversely polarized targets.

hep-ph↗
arXiv1999

NLO correction to one-particle inclusive production at high energies

Next-to-leading order correction to the one-particle inclusive cross section in the framework of high energy factorization is calculated. Numerical results for midrapidity region are compared with predictions of conventional calculations based on collinear factorization.

hep-ph↗