arXiv subjects

Iddo Eliazar

Publications and source records attributed to Iddo Eliazar.

13 recordsLinked to original sources

arXiv2022

Diversity of Sharp Restart

When applied to a stochastic process of interest, a restart protocol alters the overall statistical distribution of the process' completion time; thus, the completion-time's mean and randomness change. The explicit effect of restart on the mean is well understood, and it is known that: from a mean perspective, deterministic restart protocols -- termed sharp restart -- can out-perform any other restart protocol. However, little is known on the explicit effect of restart on randomness. This paper is the second in a duo exploring the effect of sharp restart on randomness: via a Boltzmann-Gibbs-Shannon entropy analysis in the first part, and via a diversity analysis in this part. Specifically, gauging randomness via diversity -- a measure that is intimately related to the Renyi entropy -- this paper establishes a set of universal criteria that determine: A) precisely when a sharp-restart protocol decreases/increases the diversity of completion times; B) the very existence of sharp-restart protocols that decrease/increase the diversity of completion times. Moreover, addressing jointly mean-behavior and randomness, this paper asserts and demonstrates when sharp restart has an aligned effect on the two (decreasing/increasing both), and when the effect is antithetical (decreasing one while increasing the other). The joint mean-diversity results require remarkably little information regarding the (original) statistical distributions of completion times, and are remarkably practical and easy to implement.

cond-mat.stat-mech↗

arXiv2022

Entropy of Sharp Restart

Restart has the potential of expediting or impeding the completion times of general random processes. Consequently, the issue of mean-performance takes center stage: quantifying how the application of restart on a process of interest impacts its completion-time's mean. Going beyond the mean, little is known on how restart affects stochasticity measures of the completion time. This paper is the first in a duo of studies that address this knowledge gap via: a comprehensive analysis that quantifies how sharp restart -- a keystone restart protocol -- impacts the completion-time's Boltzmann-Gibbs-Shannon entropy. The analysis establishes closed-form results for sharp restart with general timers, with fast timers (high-frequency resetting), and with slow timers (low-frequency resetting). These results share a common structure: comparing the completion-time's hazard rate to a flat benchmark -- the constant hazard rate of an exponential distribution whose entropy is equal to the completion-time's entropy. In addition, using an information-geometric approach based on Kullback-Leibler distances, the analysis establishes results that determine the very existence of timers with which the application of sharp restart decreases or increases the completion-time's entropy. Our work sheds first light on the intricate interplay between restart and randomness -- as gauged by the Boltzmann-Gibbs-Shannon entropy.

cond-mat.stat-mech↗

arXiv2021

Anomalous Diffusion: Fractional Brownian Motion vs. Fractional Ito Motion

Generalizing Brownian motion (BM), fractional Brownian motion (FBM) is a paradigmatic selfsimilar model for anomalous diffusion. Specifically, varying its Hurst exponent, FBM spans: sub-diffusion, regular diffusion, and super-diffusion. As BM, also FBM is a symmetric and Gaussian process, with a continuous trajectory, and with a stationary velocity. In contrast to BM, FBM is neither a Markov process nor a martingale, and its velocity is correlated. Based on a recent study of selfsimilar Ito diffusions, we explore an alternative selfsimilar model for anomalous diffusion: fractional Ito motion (FIM). The FIM model exhibits the same Hurst-exponent behavior as FBM, and it is also a symmetric process with a continuous trajectory. In sharp contrast to FBM, we show that FIM: is not a Gaussian process; is a Markov process; is a martingale; and its velocity is not stationary and is not correlated. On the one hand, FBM is hard to simulate, its analytic tractability is limited, and it generates only a Gaussian dissipation pattern. On the other hand, FIM is easy to simulate, it is analytically tractable, and it generates non-Gaussian dissipation patterns. Moreover, we show that FIM has an intimate linkage to diffusion in a logarithmic potential. With its compelling properties, FIM offers researchers and practitioners a highly workable analytic model for anomalous diffusion.

math.PR↗

arXiv2021

Mean-performance of Sharp Restart II: Inequality Roadmap

Restarting a deterministic process always impedes its completion. However, it is known that restarting a random process can also lead to an opposite outcome -- expediting completion. Hence, the effect of restart is contingent on the underlying statistical heterogeneity of the process' completion times. To quantify this heterogeneity we bring a novel approach to restart: the methodology of inequality indices, which is widely applied in economics and in the social sciences to measure income and wealth disparity. Using this approach we establish an `inequality roadmap' for the mean-performance of sharp restart: a whole new set of universal inequality criteria that determine when restart with sharp timers (i.e. with fixed deterministic timers) decreases/increases mean completion. The criteria are based on a host of inequality indices including Bonferroni, Gini, Pietra, and other Lorenz-curve indices; each index captures a different angle of the restart-inequality interplay. Utilizing the fact that sharp restart can match the mean-performance of any general restart protocol, we prove -- with unprecedented precision and resolution -- the validity of the following statement: restart impedes/expedites mean completion when the underlying statistical heterogeneity is low/high.

cond-mat.stat-mech↗

arXiv2020

Tail-behavior roadmap for sharp restart

Many tasks are accomplished via random processes. The completion time of such a task can be profoundly affected by restart: the occasional resetting of the task's underlying random process. Consequently, determining when restart will impede or expedite task completion is a subject of major importance. In recent years researchers explored this subject extensively, with main focus set on average behavior, i.e. on mean completion times. On the one hand, the mean approach asserts the centrality of "sharp restart" -- resetting with deterministic (fixed) timers. On the other hand, a significant drawback of the mean approach is that it provides no insight regarding tail behavior, i.e. the occurrence likelihood of extreme completion times. Addressing sharp restart, and shifting the focus from means to extremes, this paper establishes a comprehensive tail-behavior analysis of completion times. Employing the reliability-engineering notion of hazard rate, the analysis yields a set of universal results that determine -- from a tail-behavior perspective -- when sharp restart will impede or expedite task completion. The universal results are formulated in terms of explicit and highly applicable hazard-rate criteria. With the novel results at hand, a universal average-&-tail classification manual for sharp restart is devised. The manual specifies general scenarios in which -- rather counter-intuitively -- sharp restart has an opposite effect on average behavior and on tail behavior: decreasing mean completion times while dramatically increasing the likelihood of extreme completion times; and, conversely, increasing mean completion times while dramatically decreasing the likelihood of extreme completion times.

cond-mat.stat-mech↗

arXiv2020

Mean-performance of sharp restart I: Statistical roadmap

Restart is a general framework, of prime importance and wide applicability, for expediting first-passage times and completion times of general stochastic processes. Restart protocols can use either deterministic or stochastic timers. Restart protocols with deterministic timers -- "sharp restart" -- assume a principal role: if there exists a restart protocol that improves mean-performance, then there exists a sharp-restart protocol that performs as good or better. This paper, the first of a duo, presents a comprehensive mean-performance analysis of sharp restart. Using statistical methods, the analysis establishes universal criteria that determine when sharp restart improves or worsens mean-performance, i.e., decreases or increases mean first-passage/completion times. These criteria are akin to those recently discovered for the most widely applied restart protocols -- "exponential restart" -- which use exponentially-distributed timers. However, while the exponential-restart criteria cover only the case of slow timers, the sharp-restart criteria established here further cover the cases of fast, critical, and general timers; moreover, the latter criteria address the very existence of timers with which sharp restart improves or worsens mean-performance. Using the slow-timers criteria, we discover a general scenario for which: sharp restart improves mean-performance, whereas exponential restart worsens mean-performance. The potency of the novel results presented here is demonstrated by examples, and by the results' application to canonical diffusion processes.

cond-mat.stat-mech↗

arXiv2019

Gumbel Central Limit Theorem for Max-Min and Min-Max

The Max-Min and Min-Max of matrices arise prevalently in science and engineering. However, in many real-world situations the computation of the Max-Min and Min-Max is challenging as matrices are large and full information about their entries is lacking. Here we take a statistical-physics approach and establish limit-laws -- akin to the Central Limit Theorem -- for the Max-Min and Min-Max of large random matrices. The limit-laws intertwine random-matrix theory and extreme-value theory, couple the matrix-dimensions geometrically, and assert that Gumbel statistics emerge irrespective of the matrix-entries' distribution. Due to their generality and universality, as well as their practicality, these novel results are expected to have a host of applications in the physical sciences and beyond.

cond-mat.stat-mech↗

arXiv2019

Poisson-process limit-laws yield Gumbel Max-Min and Min-Max

"A chain is only as strong as its weakest link" says the proverb. But what about a collection of statistically identical chains: How long till all chains fail? The answer to this question is given by the Max-Min of a matrix whose $\left(i,j\right)$ entry is the failure time of link $j$ of chain $i$: take the minimum of each row, and then the maximum of the rows' minima. The corresponding Min-Max is obtained by taking the maximum of each column, and then the minimum of the columns' maxima. The Min-Max applies to the storage of critical data. Indeed, consider multiple backup copies of a set of critical data items, and consider the $\left(i,j\right)$ matrix entry to be the time at which item $j$ on copy $i$ is lost; then, the Min-Max is the time at which the first critical data item is lost. In this paper, we address random matrices whose entries are independent and identically distributed random variables. We establish Poisson-process limit-laws for the row's minima and for the columns' maxima. Then, we further establish Gumbel limit-laws for the Max-Min and for the Min-Max. The limit-laws hold whenever the entries' distribution has a density, and the Gumbel limit-laws yield highly applicable approximation tools and design tools for large random matrices.

cond-mat.stat-mech↗

arXiv2018

First passage under restart with branching

First passage under restart with branching is proposed as a generalization of first passage under restart. Strong motivation to study this generalization comes from the observation that restart with branching can expedite the completion of processes that cannot be expedited with simple restart; yet a sharp and quantitative formulation of this statement is still lacking. We develop a comprehensive theory of first passage under restart with branching. This reveals that two widely applied measures of statistical dispersion---the coefficient of variation and the Gini index---come together to determine how restart with branching affects the mean completion time of an arbitrary stochastic process. The universality of this result is demonstrated and its connection to extreme value theory is also pointed out and explored.

cond-mat.stat-mech↗

arXiv2013

Occupation Probabilities and Fluctuations in the Asymmetric Simple Inclusion Process

The Asymmetric Simple Inclusion Process (ASIP), a lattice-gas model of unidirectional transport and aggregation, was recently proposed as an `inclusion' counterpart of the Asymmetric Simple Exclusion Process (ASEP). In this paper we present an exact closed-form expression for the probability that a given number of particles occupies a given set of consecutive lattice sites. Our results are expressed in terms of the entries of Catalan's trapezoids --- number arrays which generalize Catalan's numbers and Catalan's triangle. We further prove that the ASIP is asymptotically governed by: (i) an inverse square root law of occupation; (ii) a square root law of fluctuation; and (iii) a Rayleigh law for the distribution of inter-exit times. The universality of these results is discussed.

cond-mat.stat-mech↗

arXiv2012

The RARE model: a generalized approach to random relaxation processes in disordered systems

This paper introduces and analyses a general statistical model, termed the RARE model, of random relaxation processes in disordered systems. The model considers excitations, that are randomly scattered around a reaction center in a general embedding space. The model's input quantities are the spatial scattering statistics of the excitations around the reaction center, and the chemical reaction rates between the excitations and the reaction center as a function of their mutual distance. The framework of the RARE model is robust, and a detailed stochastic analysis of the random relaxation processes is established. Analytic results regarding the duration and the range of the random relaxation processes, as well as the model's thermodynamic limit, are obtained in closed form. In particular, the case of power-law inputs, which turn out to yield stretched exponential relaxation patterns and asymptotically Paretian relaxation ranges, is addressed in detail.

cond-mat.stat-mech↗

arXiv2008

Power-Law Distributions: Beyond Paretian Fractality

The notion of fractality, in the context of positive-valued probability distributions, is conventionally associated with the class of Paretian probability laws. In this research we show that the Paretian class is merely one out of six classes of probability laws - all equally entitled to be ordained fractal, all possessing a characteristic power-law structure, and all being the unique fixed points of renormalizations acting on the space of positive-valued probability distributions. These six fractal classes are further shown to be one-dimensional functional projections of underlying fractal Poisson processes governed by: (i) a common elemental power-law structure; and, (ii) an intrinsic scale which can be either linear, harmonic, log-linear, or log-harmonic. This research provides a panoramic and comprehensive view of fractal distributions, backed by a unified theory of their underlying Poissonian fractals.

cond-mat.stat-mech↗

arXiv2003

On the Extreme Flights of One-Sided Levy Processes

We explore the statistical behavior of the order statistics of the flights of One-sided Levy Processes (OLPs). We begin with the study of the extreme flights of general OLPs,and then focus on the class of selfsimilar processes,investigating the following issues:(i)the inner hierarchy of the extreme flights - for example:how big is the 7th largest flight relative to the 2nd largest one?; and,(ii)the relative contribution of the extreme flights to the entire 'flight aggregate' - for example: how big is the 3rd largest flight relative to the OLP's value?. Furthermore, we show that all 'hierarchical' results obtained - but not the 'aggregate' results - are explicitly extendable to the class of OLPs with arbitrary power-law flight tails (which is far larger than the selfsimilar class).

cond-mat.stat-mech↗