Perspective Of Statistical Physics Research Articles

Introduction to Memristive Mechanisms and Models.

The increase in computational power demand led by the development of Artificial Intelligence is rapidly becoming unsustainable. New paradigms of computation, which potentially differ from digital computation, together with novel hardware architecture and devices, are anticipated to reduce the exorbitant energy demand for data-processing tasks. Memristive systems with resistive switching behavior are under intense research, given their prominent role in the fabrication of memory devices that promise the desired hardware revolution in our intensive data-driven era. They are suggested to provide the hardware substrate to scale up computational capabilities while improving their energy expenditure and speed. This work provides an orientation map for those interested in the vast topic of memristive systems with application to neuromorphic computing. We address the description of the most notable emerging devices and we illustrate models that capture the complex dynamical behavior of these systems under the dynamical-systems framework developed by Chua. We then review the memristive behavior under the perspective of statistical physics and percolation theory suited to describe fluctuations and disorder which are otherwise precluded in the dynamical-system approach. Percolation theory allows the investigation of these systems at the mesoscopic level, enabling material-independent modeling of non-linear conductance networks. We finally discuss recent and less recent successes in deep learning methods that bridge the field of physics-based and biological- inspired neuromorphic computing.

Basics of the reaction kinetics on metallic alloys.

Kinetics of heterogeneous catalytic reactions are often complicated by various factors, and from the perspective of statistical physics the development of the corresponding models is frequently challenging especially in the case of practically important alloy catalysts. To extend the basics in this area, I use a generic kinetic model of N_{2} formation inthe NO reaction with such species as CO or H_{2}. The focus is onthe reaction on the surface of a bimetallic alloy with low integral fraction of one of the metals so that the alloy is formed only at the surface. The main goal is to illustrate the specifics of the reaction kinetics and to clarify the accuracy of various approximations which are often inevitable for the analysis of such systems. The key results are as follows. (i) In the baseline one-metal case with the Langmuir-type equations, the model predicts the existence of the optimal adsorbate binding energy corresponding to the maximal reaction rate. (ii) With suitable substitution of the symbols, the equations employed for a uniform surface [item (i)] can be used for the mean-field description of reaction on the surface of a random alloy. In particular, the maximum in the dependence of the reaction rate on the binding energy is converted to the maximum with respect to the alloy composition. (iii) The reaction kinetics on the random-alloy surface have been described exactly. The corresponding maximum in the reaction rate is lower and somewhat smeared compared to that predicted in the mean-field approximation for an alloy or for the optimal monometallic catalyst. (iv) The reaction kinetics have been scrutinized also in the case of two-dimensional (2D) segregation of metal atoms at the surface in the limits of slow and rapid adsorbate diffusion between the spots. The kinetics are shown to be qualitatively different in these limits and also compared to the random-alloy case. (v) The thermodynamic criteria for 2D segregation of metal atoms at the surface have been derived as well.

Optimization by linear kinetic equations and mean-field Langevin dynamics

One of the most striking examples of the close connections between global optimization processes and statistical physics is the simulated annealing method, inspired by the famous Monte Carlo algorithm devised by Metropolis et al. in the middle of last century. In this paper, we show how the tools of linear kinetic theory allow the description of this gradient-free algorithm from the perspective of statistical physics and how convergence to the global minimum can be related to classical entropy inequalities. This analysis highlights the strong link between linear Boltzmann equations and stochastic optimization methods governed by Markov processes. Thanks to this formalism, we can establish the connections between the simulated annealing process and the corresponding mean-field Langevin dynamics characterized by a stochastic gradient descent approach. Generalizations to other selection strategies in simulated annealing that avoid the acceptance–rejection dynamic are also provided.

Entropy production of multivariate Ornstein-Uhlenbeck processes correlates with consciousness levels in the human brain.

Consciousness is supported by complex patterns of brain activity which are indicative of irreversible nonequilibrium dynamics. While the framework of stochastic thermodynamics has facilitated the understanding of physical systems of this kind, its application to infer the level of consciousness from empirical data remains elusive. We faced this challenge by calculating entropy production in a multivariate Ornstein-Uhlenbeck process fitted to Functional magnetic resonance imaging brain activity recordings. To test this approach, we focused on the transition from wakefulness to deep sleep, revealing a monotonous relationship between entropy production and the level of consciousness. Our results constitute robust signatures of consciousness while also advancing our understanding of the link between consciousness and complexity from the fundamental perspective of statistical physics.

A physical study of the LLL algorithm

This paper presents a study of the LLL algorithm from the perspective of statistical physics. Based on our experimental and theoretical results, we suggest that interpreting LLL as a sandpile model may help understand much of its mysterious behavior. In the language of physics, our work presents evidence that LLL and certain 1-d sandpile models with simpler toppling rules belong to the same universality class.This paper consists of three parts. First, we introduce sandpile models whose statistics imitate those of LLL with compelling accuracy, which leads to the idea that there must exist a meaningful connection between the two. Indeed, on those sandpile models, we are able to prove the analogues of some of the most desired statements for LLL, such as the existence of the gap between the theoretical and the experimental RHF bounds. Furthermore, we test the formulas from finite-size scaling theory (FSS) against the LLL algorithm itself, and find that they are in excellent agreement. This in particular explains and refines the geometric series assumption (GSA), and allows one to extrapolate various quantities of interest to the dimension limit. In particular, we obtain the estimate that the empirical average RHF converges to ≈1.02265 as the dimension goes to infinity.

Multiscale statistical physics of the pan-viral interactome unravels the systemic nature of SARS-CoV-2 infections

Protein–protein interaction networks have been used to investigate the influence of SARS-CoV-2 viral proteins on the function of human cells, laying out a deeper understanding of COVID–19 and providing ground for applications, such as drug repurposing. Characterizing molecular (dis)similarities between SARS-CoV-2 and other viral agents allows one to exploit existing information about the alteration of key biological processes due to known viruses for predicting the potential effects of this new virus. Here, we compare the novel coronavirus network against 92 known viruses, from the perspective of statistical physics and computational biology. We show that regulatory spreading patterns, physical features and enriched biological pathways in targeted proteins lead, overall, to meaningful clusters of viruses which, across scales, provide complementary perspectives to better characterize SARS-CoV-2 and its effects on humans. Our results indicate that the virus responsible for COVID–19 exhibits expected similarities, such as to Influenza A and Human Respiratory Syncytial viruses, and unexpected ones with different infection types and from distant viral families, like HIV1 and Human Herpes virus. Taken together, our findings indicate that COVID–19 is a systemic disease with potential effects on the function of multiple organs and human body sub-systems.

An Applied Component Modeling to the Irreversibility from a New Configurationally Perspective of the Statistical Physics

From the perspective of statistical physics (Boltzmann equation), configurational entropy can be calculated using the study of the microstates of the system. When a physical process is performed, identifying the entropy production can be used to investigate the irreversibility, but from the perspective of the Boltzmann equation, to study entropy production, both all microstates and macrostates must be studied. Therefore, a very large volume of calculations will be needed. In this report, using a new innovative energy structure equation, a new macroscopic component modeling is extracted to investigate the configurational irreversibility. To investigate the irreversibility in physical systems, the energy structure equation of the system can be studied in different paths. During performing a physical process, some activated energy components related to the reversible process and remain will be related to the irreversible process. In this report, also using a quasi-statistical approach, the structure of irreversible components is studied. When macroscopic energy components are the base of the equations, a very large volume of the needed calculations will be less than Boltzmann equation and in fact, studying all particles isn’t needed, but it is enough that a few macroscopic components to be investigated. Also, considering the theories of dissipated energy, the extracted equations have the same base as the different formulations of the second law of thermodynamics.

The Car-Following Model Based on Fuzzy Inference Controller

Many existing researches mostly establish a theoretical car-following model from a perspective of statistical physics, but the vehicle driver is the decision-maker of driving behaviors. When modeling the micro-behavior of the vehicle, the driver’s driving intention and decision-making basis need to be considered. In order to consider human factors more in microscopic traffic simulation, this paper uses fuzzy inference technology as the basis, selects the headway distance, ideal speed difference, and reasonable spacing difference as the input variables of the control system, considering the free running and car-following state of the vehicle, constructs fuzzy rules and establishes a car-following model.

Experimentally Accessible Witnesses of Many-Body Localization

The phenomenon of many-body localized (MBL) systems has attracted significant interest in recent years, for its intriguing implications from a perspective of both condensed-matter and statistical physics: they are insulators even at non-zero temperature and fail to thermalize, violating expectations from quantum statistical mechanics. What is more, recent seminal experimental developments with ultra-cold atoms in optical lattices constituting analog quantum simulators have pushed many-body localized systems into the realm of physical systems that can be measured with high accuracy. In this work, we introduce experimentally accessible witnesses that directly probe distinct features of MBL, distinguishing it from its Anderson counterpart. We insist on building our toolbox from techniques available in the laboratory, including on-site addressing, super-lattices, and time-of-flight measurements, identifying witnesses based on fluctuations, density–density correlators, densities, and entanglement. We build upon the theory of out of equilibrium quantum systems, in conjunction with tensor network and exact simulations, showing the effectiveness of the tools for realistic models.

Statistical physics of liquid brains.

Liquid neural networks (or 'liquid brains') are a widespread class of cognitive living networks characterized by a common feature: the agents (ants or immune cells, for example) move in space. Thus, no fixed, long-term agent-agent connections are maintained, in contrast with standard neural systems. How is this class of systems capable of displaying cognitive abilities, from learning to decision-making? In this paper, the collective dynamics, memory and learning properties of liquid brains is explored under the perspective of statistical physics. Using a comparative approach, we review the generic properties of three large classes of systems, namely: standard neural networks (solid brains), ant colonies and the immune system. It is shown that, despite their intrinsic physical differences, these systems share key properties with standard neural systems in terms of formal descriptions, but strongly depart in other ways. On one hand, the attractors found in liquid brains are not always based on connection weights but instead on population abundances. However, some liquid systems use fluctuations in ways similar to those found in cortical networks, suggesting a relevant role for criticality as a way of rapidly reacting to external signals. This article is part of the theme issue 'Liquid brains, solid brains: How distributed cognitive architectures process information'.

Differences in the collective behaviour of the genes in the cancerous and normal cell from the perspective of inverse statistical physics

Differences in the collective behaviour of the genes in the cancerous and normal cell from the perspective of inverse statistical physics

Self-similar characteristics of single nucleotide polymorphisms in the rice genome

With single nucleotide polymorphism (SNP) data from the 3,000 rice genome project, we investigate the mutational characteristics of the rice genome from the perspective of statistical physics. From the frequency distributions of the space between adjacent SNPs, we present evidence that SNPs are not spaced randomly, but clustered across the genome. The clustering property is related to a long-range correlation in SNP locations, suggesting that a mutation occurring in a locus may affect other mutations far away along the sequence in a chromosome. In addition, the reliability of the existence of the long-range correlation is supported by the agreement between the results of two independent analysis methods. The highly-skewed and long-tailed distribution of SNP spaces is further characterized by a multi-fractal, showing that SNP spaces possess a rich structure of a statistical self-similarity. These results can be used for an optimal design of a microarray assay and a primer, as well as for genotyping quality control.

Breaking news dissemination in the media via propagation behavior based on complex network theory

The diffusion of breaking news largely relies on propagation behaviors in the media. The tremendous and intricate propagation relationships in the media form a complex network. An improved understanding of breaking news diffusion characteristics can be obtained through the complex network research. Drawing on the news data of Bohai Gulf oil spill event from June 2011 to May 2014, we constructed a weighted and directed complex network in which media are set as nodes, the propagation relationships as edges and the propagation times as the weight of the edges. The primary results show (1) the propagation network presents small world feature, which means relations among media are close and breaking news originating from any node can spread rapidly; (2) traditional media and official websites are the typical sources for news propagation, while business portals are news collectors and spreaders; (3) the propagation network is assortative and the group of core media facilities the spread of breaking news faster; (4) for online media, news originality factor become less important to propagation behaviors. This study offers a new insight to explore information dissemination from the perspective of statistical physics and is beneficial for utilizing the public opinion in a positive way.

Resistive switching phenomena: A review of statistical physics approaches

Resistive switching (RS) phenomena are reversible changes in the metastable resistance state induced by external electric fields. After discovery ∼50 years ago, RS phenomena have attracted great attention due to their potential application in next-generation electrical devices. Considerable research has been performed to understand the physical mechanisms of RS and explore the feasibility and limits of such devices. There have also been several reviews on RS that attempt to explain the microscopic origins of how regions that were originally insulators can change into conductors. However, little attention has been paid to the most important factor in determining resistance: how conducting local regions are interconnected. Here, we provide an overview of the underlying physics behind connectivity changes in highly conductive regions under an electric field. We first classify RS phenomena according to their characteristic current–voltage curves: unipolar, bipolar, and threshold switchings. Second, we outline the microscopic origins of RS in oxides, focusing on the roles of oxygen vacancies: the effect of concentration, the mechanisms of channel formation and rupture, and the driving forces of oxygen vacancies. Third, we review RS studies from the perspective of statistical physics to understand connectivity change in RS phenomena. We discuss percolation model approaches and the theory for the scaling behaviors of numerous transport properties observed in RS. Fourth, we review various switching-type conversion phenomena in RS: bipolar-unipolar, memory-threshold, figure-of-eight, and counter-figure-of-eight conversions. Finally, we review several related technological issues, such as improvement in high resistance fluctuations, sneak-path problems, and multilevel switching problems.

Power Law Behavior and Self-Similarity in Modern Industrial Accidents

Advances in technology have produced more and more intricate industrial systems, such as nuclear power plants, chemical centers and petroleum platforms. Such complex plants exhibit multiple interactions among smaller units and human operators, rising potentially disastrous failure, which can propagate across subsystem boundaries. This paper analyzes industrial accident data-series in the perspective of statistical physics and dynamical systems. Global data is collected from the Emergency Events Database (EM-DAT) during the time period from year 1903 up to 2012. The statistical distributions of the number of fatalities caused by industrial accidents reveal Power Law (PL) behavior. We analyze the evolution of the PL parameters over time and observe a remarkable increment in the PL exponent during the last years. PL behavior allows prediction by extrapolation over a wide range of scales. In a complementary line of thought, we compare the data using appropriate indices and use different visualization techniques to correlate and to extract relationships among industrial accident events. This study contributes to better understand the complexity of modern industrial accidents and their ruling principles.

EMPLOYMENT, PRODUCTION AND CONSUMPTION WITH RANDOM UPDATE: NON-EQUILIBRIUM STATIONARY STATE EQUATIONS

In this work, we investigate the Model of Employment, Production and Consumption, as introduced in a series of papers by I. Wright [1–3] from the perspective of statistical physics, and we focus on the presence of equilibrium. The model itself belongs to the class of multi-agent computational models, which aim to explain macro-economic behavior using explicit micro-economic interactions.<br />Based on the mean-field approximation, we form the Fokker-Plank equation(s) and then formulate conditions forming the stationary solution, which results in a system of non-linear integral-differential equations. This approximation then allows the presence of non-equilibrium stationary states, where the model is a mixed additive-multiplicative model.

TEMPLATE-DIRECTED BIOPOLYMERIZATION: TAPE-COPYING TURING MACHINES

DNA, RNA and proteins are among the most important macromolecules in a living cell. These molecules are polymerized by molecular machines. These natural nano-machines polymerize such macromolecules, adding one monomer at a time, using another linear polymer as the corresponding template. The machine utilizes input chemical energy to move along the template which also serves as a track for the movements of the machine. In the Alan Turing year 2012, it is worth pointing out that these machines are "tape-copying Turing machines". We review the operational mechanisms of the polymerizer machines and their collective behavior from the perspective of statistical physics, emphasizing their common features in spite of the crucial differences in their biological functions. We also draw the attention of the physics community to another class of modular machines that carry out a different type of template-directed polymerization. We hope this review will inspire new kinetic models for these modular machines.

Dynamics of traffic flow with real-time traffic information.

We studied dynamics of traffic flow with real-time information provided. Provision of the real-time traffic information based on advancements in telecommunication technology is expected to facilitate the efficient utilization of available road capacity. This system has a potentiality of not only engineering for road usage but also the science of complexity series. In the system, the information plays a role of feedback connecting microscopic and macroscopic phenomena beyond the hierarchical structure of statistical physics. In this paper, we tried to clarify how the information works in a network of traffic flow from the perspective of statistical physics. The dynamical feature of the traffic flow is abstracted by a contrastive study between the nonequilibrium statistical physics and a computer simulation based on cellular automaton. We found that the information disrupts the local equilibrium of traffic flow by a characteristic dissipation process due to interaction between the information and individual vehicles. The dissipative structure was observed in the time evolution of traffic flow driven far from equilibrium as a consequence of the breakdown of the local-equilibrium hypothesis.

Four lectures on the physics of crystal growth

Several aspects of the theory of epitaxial crystal growth from atomic or molecular beams are developed from the perspective of statistical physics. Lectures are devoted to the rate equation theory of two-dimensional nucleation and its limitations; the growth of multilayer wedding cakes in the presence of strong step edge barriers; the continuum theory of mound coarsening; and growth-induced step meandering on vicinal surfaces.