arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.03481cond-mat.stat-mechphysics.bio-phq-bio.PE
收藏

非马尔可夫种群动力学中从亚稳态的逃逸

Escape from a metastable state in non-Markovian population dynamics

Ohad Vilk, Michael Assaf

AI总结:

本文研究含厚尾等待时间记忆的非马尔可夫单种群随机动力学,用 WKB 方法分析有限均值情形的亚稳态逃逸,并揭示无限均值下双稳态系统的两种标度机制。

摘要:

我们研究非马尔可夫单种群随机模型中的长时间动力学,其中一个或多个反应被建模为等待时间服从厚尾非指数分布的随机过程,用以模拟长期记忆。我们聚焦于三个典型例子:基因切换、种群建立和种群灭绝,它们均具有非指数产生率。该系统在两种机制下得到研究。在第一种机制中,等待时间分布具有有限均值。此时,系统在长时间尺度上趋近于一个(准)稳态,我们为这些非马尔可夫系统发展了一种通用的 WKB 方法。我们推导了随机动力学中平均种群规模和从亚稳态逃逸的平均时间的显式结果。在这一范围内,我们揭示出,对于足够强的记忆,系统中可以出现由记忆诱导的(亚)稳态。在第二种机制中,假设等待时间分布具有无限均值。此时,对于双稳态系统,我们发现两种不同的标度机制,它们被一个指数长时间分隔,而该时间可能强烈依赖于系统的初始条件。

英文摘要:

We study the long-time dynamics in non-Markovian single-population stochastic models, where one or more reactions are modelled as a stochastic process with a fat-tailed non-exponential distribution of waiting times, mimicking long-term memory. We focus on three prototypical examples: genetic switching, population establishment and population extinction, all with non-exponential production rates. The system is studied in two regimes. In the first, the distribution of waiting times has a finite mean. Here, the system approaches a (quasi)stationary steady state at long times, and we develop a general WKB approach for these non-Markovian systems. We derive explicit results for the mean population size and mean escape time from the metastable state of the stochastic dynamics. In this realm, we reveal that for sufficiently strong memory, a memory-induced (meta)stable state can emerge in the system. In the second regime, the waiting time distribution is assumed to have an infinite mean. Here, for bistable systems we find two distinct scaling regimes, separated by an exponentially long time which may strongly depend on the initial conditions of the system.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑