Probability
New submissions
[ showing up to 2000 entries per page: fewer | more ]
New submissions for Thu, 25 Apr 24
- [1] arXiv:2404.15651 [pdf, ps, other]
-
Title: Sampling from Spherical Spin Glasses in Total Variation via Algorithmic Stochastic LocalizationComments: 107 pagesSubjects: Probability (math.PR); Disordered Systems and Neural Networks (cond-mat.dis-nn); Mathematical Physics (math-ph)
We consider the problem of algorithmically sampling from the Gibbs measure of a mixed $p$-spin spherical spin glass. We give a polynomial-time algorithm that samples from the Gibbs measure up to vanishing total variation error, for any model whose mixture satisfies $$\xi''(s) < \frac{1}{(1-s)^2}, \qquad \forall s\in [0,1).$$ This includes the pure $p$-spin glasses above a critical temperature that is within an absolute ($p$-independent) constant of the so-called shattering phase transition. Our algorithm follows the algorithmic stochastic localization approach introduced in (Alaoui, Montanari, Sellke, 20022). A key step of this approach is to estimate the mean of a sequence of tilted measures. We produce an improved estimator for this task by identifying a suitable correction to the TAP fixed point selected by approximate message passing (AMP). As a consequence, we improve the algorithm's guarantee over previous work, from normalized Wasserstein to total variation error. In particular, the new algorithm and analysis opens the way to perform inference about one-dimensional projections of the measure.
- [2] arXiv:2404.15779 [pdf, ps, other]
-
Title: Divergence metrics in the study of Markov and hidden Markov processesSubjects: Probability (math.PR)
This paper is in two parts. In the first part of this paper, formulae for f-divergence for a continuous Markov process are reviewed. Applications of these formulae are described for the problems of stochastic stability, second law of thermodynamics, and non-equilibrium extensions thereof. The first part sets the stage for considering the f-divergence for hidden Markov processes which is the focus of the second part. For hidden Markov processes, again based on the use of f-divergence, analyses of filter stability and stochastic thermodynamics are described. The latter is used to illustrate the Maxwell demon for an over-damped Langevin model with white noise type observations. The expository nature of the paper combined with consistent formalism for both Markov and hidden Markov processes is written to be useful for researchers working on related themes in disparate areas.
- [3] arXiv:2404.16021 [pdf, other]
-
Title: Critical beta-splitting, via contractionAuthors: Brett KolesnikSubjects: Probability (math.PR); Combinatorics (math.CO)
The critical beta-splitting tree, introduced by Aldous, is a Markov branching phylogenetic tree of poly-logarithmic height. Recently, by a technical analysis, Aldous and Pittel proved, amongst other results, a central limit theorem for the height $H_n$ of a random leaf.
We give an alternative proof, via contraction methods for random recursive structures. These techniques were developed by Neininger and R\"uschendorf, motivated by Pittel's article "Normal convergence problem? Two moments and a recurrence may be the clues." Aldous and Pittel estimated the first two moments of $H_n$, with great precision. We show that a limit theorem follows, and bound the distance from normality. - [4] arXiv:2404.16034 [pdf, ps, other]
-
Title: Central limit theorems associated with the hierarchical Dirichlet processComments: 30 pages. Comments welcomeSubjects: Probability (math.PR)
The Dirichlet process is a discrete random measure specified by a concentration parameter and a base distribution, and is used as a prior distribution in Bayesian nonparametrics. The hierarchical Dirichlet process generalizes the Dirichlet process by randomizing the base distribution through a draw from another Dirichlet process. It is motivated by the study of groups of clustered data, where the group specific Dirichlet processes are linked through an intergroup Dirichlet process. Focusing on an individual group, the hierarchical Dirichlet process is a discrete random measure whose weights have stronger dependence than the weights of the Dirichlet process. In this paper, we study the asymptotic behavior of the power sum symmetric polynomials for the vector of weights of the hierarchical Dirichlet process when the corresponding concentration parameters tend to infinity. We establish central limit theorems and obtain explicit representations for the asymptotic variances, with the latter clearly showing the impact of the hierarchical structure. These objects are closely related to the homozygosity in population genetics, the Simpson diversity index in ecology, and the Herfindahl-Hirschman index in economics.
Cross-lists for Thu, 25 Apr 24
- [5] arXiv:2404.15434 (cross-list from math.CA) [pdf, ps, other]
-
Title: Fractal uncertainty principle for random Cantor setsComments: 18 pages, 3 figuresSubjects: Classical Analysis and ODEs (math.CA); Probability (math.PR)
We continue our investigation of the fractal uncertainty principle (FUP) for random fractal sets. In the prequel (arXiv:2107.08276), we considered the Cantor sets in the discrete setting with alphabets randomly chosen from a base of digits so the dimension d is in (0,2/3). We proved that, with overwhelming probability, the FUP with an exponent >=1/2-3d/4- holds for these discrete Cantor sets with random alphabets.
In this sequel, we construct random Cantor sets with dimension d in (0,2/3) in R via a different random procedure from the one in the prequel. We prove that, with overwhelming probability, the FUP with an exponent >=1/2-3d/4- holds. The proof follows from establishing a Fourier decay estimate of the corresponding random Cantor measures, which is in turn based on a concentration of measure phenomenon in an appropriate probability space for the random Cantor sets. - [6] arXiv:2404.15483 (cross-list from cs.GT) [pdf, other]
-
Title: Strategy Complexity of Büchi Objectives in Concurrent Stochastic GamesSubjects: Computer Science and Game Theory (cs.GT); Probability (math.PR)
We study 2-player concurrent stochastic B\"uchi games on countable graphs. Two players, Max and Min, seek respectively to maximize and minimize the probability of visiting a set of target states infinitely often. We show that there always exist $\varepsilon$-optimal Max strategies that use just a step counter plus 1 bit of public memory. This upper bound holds for all countable graphs, but it is a new result even for the special case of finite graphs. The upper bound is tight in the sense that Max strategies that use just a step counter, or just finite memory, are not sufficient even on finite game graphs.
The upper bound is a consequence of a slightly stronger new result: $\varepsilon$-optimal Max strategies for the combined B\"uchi and Transience objective require just 1 bit of public memory (but cannot be memoryless). Our proof techniques also yield a closely related result, that $\varepsilon$-optimal Max strategies for the Transience objective alone (which is only meaningful in infinite graphs) can be memoryless. - [7] arXiv:2404.15490 (cross-list from math-ph) [pdf, other]
-
Title: Fock space of local fields of the discrete GFF and its scaling limit bosonic CFTComments: 84 pages, 14 figures, comments welcome!Subjects: Mathematical Physics (math-ph); Probability (math.PR)
To connect conformal field theories (CFT) to probabilistic lattice models, recent works [HKV22, Ada23] have introduced a novel definition of local fields of the lattice models. Local fields in this picture are probabilistically concrete: they are built from random variables in the model. The key insight is that discrete complex analysis ideas allow to equip the space of local fields with the main structure of a CFT: a representation of the Virasoro algebra.
In this article, for the first time, we fully analyze the structure of the space of local fields of a lattice model as a representation, and use this to establish a one-to-one correspondence between the local fields of a lattice model and those of a conformal field theory. The CFT we consider is probabilistically realized in terms of the gradient of the Gaussian Free Field (GFF). Its space of local fields is just a bosonic Fock space for two chiral symmetry algebras. The corresponding lattice model is the discrete Gaussian Free Field. Our first main result is that the space of local fields of polynomials in the gradient of the discrete GFF is isomorphic to the Fock space. Notably, local fields in this setup make sense with both Dirichlet and Neumann boundary conditions. Our second main result is that with the appropriate renormalization, correlation functions of local fields of the discrete GFF converge in the scaling limit to the correlation functions of the CFT. The renormalization needed is, conceptually correctly, according to the eigenvalue of the Virasoro generator $L_0 + \bar{L}_0$ on the local field. - [8] arXiv:2404.15623 (cross-list from cs.NI) [pdf, other]
-
Title: Characterizing the Age of Information with Multiple Coexisting Data StreamsSubjects: Networking and Internet Architecture (cs.NI); Probability (math.PR)
In this paper we analyze the distribution of the Age of Information (AoI) of a tagged data stream sharing a processor with a set of other data streams. We do so in the highly general setting in which the interarrival times pertaining to the tagged stream can have any distribution, and also the service times of both the tagged stream and the background stream are generally distributed. The packet arrival times of the background process are assumed to constitute a Poisson process, which is justified by the fact that it typically is a superposition of many relatively homogeneous streams. The first major contribution is that we derive an expression for the Laplace-Stieltjes transform of the AoI in the resulting GI+M/GI+GI/1 model. Second, we use stochastic ordering techniques to identify tight stochastic bounds on the AoI. In addition, when approximating the tagged stream's inter-generation times through a phase-type distribution (which can be done at any precision), we present a computational algorithm for the mean AoI. As illustrated through a sequence of numerical experiments, the analysis enables us to assess the impact of background traffic on the AoI of the tagged stream.
- [9] arXiv:2404.15645 (cross-list from math.DG) [pdf, ps, other]
-
Title: Spectral Gap Estimates on Conformally Flat ManifoldsComments: 23 pagesSubjects: Differential Geometry (math.DG); Analysis of PDEs (math.AP); Probability (math.PR); Spectral Theory (math.SP)
The fundamental gap is the difference between the first two Dirichlet eigenvalues of a Schr\"odinger operator (and the Laplacian, in particular). For horoconvex domains in hyperbolic space, Nguyen, Stancu and Wei conjectured that it is possible to obtain a lower bound on the fundamental gap in terms of the diameter of the domain and the dimension [IMRN2022]. In this article, we prove this conjecture by establishing conformal log-concavity estimates for the first eigenfunction. This builds off earlier work by the authors and Saha as well as recent work by Cho, Wei and Yang. We also prove spectral gap estimates for a more general class of problems on conformally flat manifolds and investigate the relationship between the gap and the inradius. For example, we establish gap estimates for domains in $\mathbb{S}^1 \times \mathbb{S}^{N-1}$ which are convex with respect to the universal affine cover
- [10] arXiv:2404.15725 (cross-list from math.AP) [pdf, other]
-
Title: Local convergence rates for Wasserstein gradient flows and McKean-Vlasov equations with multiple stationary solutionsSubjects: Analysis of PDEs (math.AP); Functional Analysis (math.FA); Probability (math.PR)
Non-linear versions of log-Sobolev inequalities, that link a free energy to its dissipation along the corresponding Wasserstein gradient flow (i.e. corresponds to Polyak-Lojasiewicz inequalities in this context), are known to provide global exponential long-time convergence to the free energy minimizers, and have been shown to hold in various contexts. However they cannot hold when the free energy admits critical points which are not global minimizers, which is for instance the case of the granular media equation in a double-well potential with quadratic attractive interaction at low temperature. This work addresses such cases, extending the general arguments when a log-Sobolev inequality only holds locally and, as an example, establishing such local inequalities for the granular media equation with quadratic interaction either in the one-dimensional symmetric double-well case or in higher dimension in the low temperature regime. The method provides quantitative convergence rates for initial conditions in a Wasserstein ball around the stationary solutions. The same analysis is carried out for the kinetic counterpart of the gradient flow, i.e. the corresponding Vlasov-Fokker-Planck equation. The local exponential convergence to stationary solutions for the mean-field equations, both elliptic and kinetic, is shown to induce for the corresponding particle systems a fast (i.e. uniform in the number or particles) decay of the particle system free energy toward the level of the non-linear limit.
- [11] arXiv:2404.15853 (cross-list from cond-mat.stat-mech) [pdf, other]
-
Title: Exact propagators of one-dimensional self-interacting random walksSubjects: Statistical Mechanics (cond-mat.stat-mech); Probability (math.PR)
Self-interacting random walks (SIRWs) show long-range memory effects that result from the interaction of the random walker at time $t$ with the territory already visited at earlier times $t'<t$. This class of non-Markovian random walks has applications in contexts as diverse as foraging theory, the behaviour of living cells, and even machine learning. Despite this importance and numerous theoretical efforts, the propagator, which is the distribution of the walker's position and arguably the most fundamental quantity to characterize the process, has so far remained out of reach for all but a single class of SIRW. Here we fill this gap and provide an exact and explicit expression for the propagator of two important classes of SIRWs, namely, the once-reinforced random walk and the polynomially self-repelling walk. These results give access to key observables, such as the diffusion coefficient, which so far had not been determined. We also uncover an inherently non-Markovian mechanism that tends to drive the walker away from its starting point.
Replacements for Thu, 25 Apr 24
- [12] arXiv:2105.13925 (replaced) [pdf, ps, other]
-
Title: Conformally invariant random fields, quantum Liouville measures, and random Paneitz operators on Riemannian manifolds of even dimensionComments: New notion of `singular' h, now Sobolev function rather than distribution. Detailed discussion on plain vs. adjusted LQG measure. Additional results on Polyakov-Liouville measureSubjects: Probability (math.PR)
- [13] arXiv:2107.01878 (replaced) [pdf, ps, other]
-
Title: Percolation transition for random forests in $d\geq 3$Comments: Accepted versionSubjects: Probability (math.PR); Mathematical Physics (math-ph); Combinatorics (math.CO)
- [14] arXiv:2212.03705 (replaced) [pdf, other]
-
Title: Aggregate Markov models in life insurance: properties and valuationSubjects: Probability (math.PR)
- [15] arXiv:2304.09127 (replaced) [pdf, other]
-
Title: Survival and complete convergence for a branching annihilating random walkSubjects: Probability (math.PR)
- [16] arXiv:2307.07619 (replaced) [pdf, other]
-
Title: Stochastic dynamics and the Polchinski equation: an introductionSubjects: Probability (math.PR); Mathematical Physics (math-ph); Functional Analysis (math.FA)
- [17] arXiv:2308.05639 (replaced) [pdf, ps, other]
-
Title: Distributional properties of jumps of multi-type CBI processesComments: 55 pages. Title has been changedSubjects: Probability (math.PR)
- [18] arXiv:2401.12721 (replaced) [pdf, other]
-
Title: Wasserstein Diffusion on Multidimensional SpacesAuthors: Karl-Theodor SturmComments: New result on Large Deviation Principle (Thm. 3.9 + 3.10); corrected proof for Lemma 2.6Subjects: Probability (math.PR); Functional Analysis (math.FA); Metric Geometry (math.MG)
- [19] arXiv:2402.15130 (replaced) [pdf, ps, other]
-
Title: Diffusion Processes on $p$-Wasserstein Space over Banach SpaceComments: Acknowledgments added; Application of Sect. 3.2 revised; Typos correctedSubjects: Probability (math.PR)
- [20] arXiv:2403.05372 (replaced) [pdf, ps, other]
-
Title: Limit Laws for Critical Dispersion on Complete GraphsComments: 37 pages 2 figuresSubjects: Probability (math.PR); Discrete Mathematics (cs.DM); Combinatorics (math.CO)
- [21] arXiv:2403.07618 (replaced) [pdf, other]
-
Title: Markov Chain Aggregation with Error Bounds on Transient DistributionsComments: 36 pagesSubjects: Probability (math.PR)
- [22] arXiv:2112.10953 (replaced) [pdf, other]
-
Title: An adaptation of InfoMap to absorbing random walks using absorption-scaled graphsSubjects: Social and Information Networks (cs.SI); Machine Learning (cs.LG); Probability (math.PR); Adaptation and Self-Organizing Systems (nlin.AO); Physics and Society (physics.soc-ph)
- [23] arXiv:2303.02079 (replaced) [pdf, other]
-
Title: Insights from number theory into the critical Kauffman model with connectivity oneComments: 15 pages, 3 figuresSubjects: Molecular Networks (q-bio.MN); Disordered Systems and Neural Networks (cond-mat.dis-nn); Probability (math.PR)
- [24] arXiv:2404.06023 (replaced) [pdf, other]
-
Title: Prelimit Coupling and Steady-State Convergence of Constant-stepsize Nonsmooth Contractive SAComments: ACM SIGMETRICS 2024. 71 pages, 3 figuresSubjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC); Probability (math.PR)
- [25] arXiv:2404.15191 (replaced) [pdf, ps, other]
-
Title: Convergence of martingales via enriched dagger categoriesComments: 64 pagesSubjects: Category Theory (math.CT); Probability (math.PR)
[ showing up to 2000 entries per page: fewer | more ]
Disable MathJax (What is MathJax?)
Links to: arXiv, form interface, find, math, recent, 2404, contact, help (Access key information)