arXiv subjects

G. A. Alves

Publications and source records attributed to G. A. Alves.

At least 19 recordsLinked to original sources

arXiv2026

Unsupervised Machine Learning of the Contact Process

We investigate how unsupervised machine-learning methods can characterize the active-absorbing phase transition in the contact process in one and two spatial dimensions. Our analysis focuses on principal component analysis (PCA) and variational autoencoders (VAEs), and we show that direct applications of these methods encounter a central difficulty: contact process configurations are intrinsically positive-definite. In the PCA case, standard centering procedures fail to isolate the order parameter in the leading component. To address this, we augment the data with sign-reversed configurations, creating a balanced dataset with zero mean while preserving the density information required to describe the transition. This preprocessing allows the dominant principal component to recover the order parameter and reproduce the expected finite-size scaling near criticality. For VAEs, we find a related limitation: a fixed Gaussian prior does not provide sufficient latent regularization across the full range of control parameters and system sizes for positive-definite data. We overcome this by adopting a heteroscedastic Gaussian prior that adapts to the control parameter, leading to a substantial improvement in the finite-size scaling of reconstruction-based observables near the critical threshold. Taken together, these results highlight a key lesson: the success of machine learning in phase-transition problems does not rely on algorithmic complexity alone, but on tailoring the representation to respect the physical structure of the system, including its intrinsic constraints and symmetries.

cond-mat.stat-mech↗

arXiv2026

Kinetic Flat-Histogram Simulations of Non-Equilibrium Stochastic Processes with Continuous and Discontinuous Phase Transitions

As far as we know, there is no flat-histogram algorithm to sample the stationary distribution of non-equilibrium stochastic processes. The present work addresses this gap by introducing a generalization of the Wang-Landau algorithm, applied to non-equilibrium stochastic processes with local transitions. The main idea is to sample macroscopic states using a kinetic Monte Carlo algorithm to generate trial moves, which are accepted or rejected with a probability that depends inversely on the stationary distribution. The stationary distribution is refined through the simulation by a modification factor, leading to convergence toward the true stationary distribution. A visitation histogram is also accumulated, and the modification factor is updated when the histogram satisfies a flatness condition. The stationary distribution is obtained in the limit where the modification factor reaches a threshold value close to unity. To test the algorithm, we compare simulation results for several stochastic processes with theoretically known behavior. In addition, results from the kinetic flat-histogram algorithm are compared with standard exact stochastic simulations. We show that the kinetic flat-histogram algorithm can be applied to phase transitions in stochastic processes with bistability, which describe a wide range of phenomena such as epidemic spreading, population growth, chemical reactions, and consensus formation. With some adaptations, the kinetic flat-histogram algorithm can also be applied to stochastic models on lattices and complex networks.

cond-mat.stat-mech↗

arXiv2025

Supervised and Unsupervised Deep Learning Applied to the Majority Vote Model

We employ deep learning techniques to investigate the critical properties of the continuous phase transition in the majority vote model. In addition to deep learning, principal component analysis is utilized to analyze the transition. For supervised learning, dense neural networks are trained on spin configuration data generated via the kinetic Monte Carlo method. Using independently simulated configuration data, the neural network accurately identifies the critical point on both square and triangular lattices. Classical unsupervised learning with principal component analysis reproduces the magnetization and enables estimation of critical exponents, typically obtained via Monte Carlo importance sampling. Furthermore, deep unsupervised learning is performed using variational autoencoders, which reconstruct input spin configurations and generate artificial outputs. The autoencoders detect the phase transition through the loss function, quantifying the preservation of essential data features. We define a correlation function between the real and reconstructed data, and find that this correlation function is universal at the critical point. Variational autoencoders also serve as generative models, producing artificial spin configurations.

cond-mat.stat-mech↗

arXiv2025

Deep Learning of the Biswas-Chatterjee-Sen Model

We investigate the critical properties of kinetic continuous opinion dynamics using deep learning techniques. The system consists of $N$ continuous spin variables in the interval $[-1,1]$. Dense neural networks are trained on spin configuration data generated via kinetic Monte Carlo simulations, accurately identifying the critical point on both square and triangular lattices. Classical unsupervised learning with principal component analysis reproduces the magnetization and allows estimation of critical exponents. Additionally, variational autoencoders are implemented to study the phase transition through the loss function, which behaves as an order parameter. A correlation function between real and reconstructed data is defined and found to be universal at the critical point.

cond-mat.stat-mech↗

arXiv2025

CMS RPC Non-Physics Event Data Automation Ideology

This paper presents a streamlined framework for real-time processing and analysis of condition data from the CMS experiment Resistive Plate Chambers (RPC). Leveraging data streaming, it uncovers correlations between RPC performance metrics, like currents and rates, and LHC luminosity or environmental conditions. The Java-based framework automates data handling and predictive modeling, integrating extensive datasets into synchronized, query-optimized tables. By segmenting LHC operations and analyzing larger virtual detector objects, the automation enhances monitoring precision, accelerates visualization, and provides predictive insights, revolutionizing RPC performance evaluation and future behavior modeling.

hep-ex↗

arXiv2024

Social dilemmas, network reciprocity and the small-world property

We revisit two evolutionary game theory models, namely the Prisoner and the Snowdrift dilemmas, on top of small-world networks. These dynamics on networked populations (individuals occupying nodes of a graph) are mainly concerning on the competition between to cooperate or to defect, by allowing some process of revision of strategies. Cooperators avoid defectors by forming clusters in a process known as network reciprocity. This defense strategy is based on the fact that any individual interact only with its nearest neighbors. The minimum cluster, in turn, is formed by a set of three completely connected nodes and the bulk of these triplets is associated with the transitivity property of a network. Particularly, we show that the transitivity increases eventually assuming a constant behavior when observed as a function of the number of contacts of an individual. We investigate the influence of the network reciprocity on that transitivity increasing regime on the promotion of a cooperative behavior. The dynamics on small-world networks are compared with those random regular, and annealed networks, the later typically studied as the well-mixed approach. We observe that the Snowdrift Game converge to an annealed scenario as randonness and coordination number increase, whereas the Prisoner's Dilemma becomes more severe against the cooperative behavior under the regime of an increasing network reciprocity.

physics.soc-ph↗

arXiv2024

Update of the Brazilian Participation in the Next-Generation Collider Experiments

This proposal outlines the future plans of the Brazilian High-Energy Physics (HEP) community for upcoming collider experiments. With the construction of new particle colliders on the horizon and the ongoing operation of the High-Luminosity LHC, several research groups in Brazil have put forward technical proposals, covering both hardware and software contributions, as part of the Brazilian contribution to the global effort. The primary goal remains to foster a unified effort within the Brazilian HEP community, optimizing resources and expertise to deliver a high-impact contribution to the international HEP community.

hep-ex↗

arXiv2024

Generalized Diffusive Epidemic Process with Permanent Immunity in Two Dimensions

We introduce the generalized diffusive epidemic process, which is a metapopulation model for an epidemic outbreak where a non-sedentary population of walkers can jump along lattice edges with diffusion rates $D_S$ or $D_I$ if they are susceptible or infected, respectively, and recovered individuals possess permanent immunity. Individuals can be contaminated with rate $μ_c$ if they share the same lattice node with an infected individual and recover with rate $μ_r$, being removed from the dynamics. Therefore, the model does not have the conservation of the active particles composed of susceptible and infected individuals. The reaction-diffusion dynamics are separated into two stages: (i) Brownian diffusion, where the particles can jump to neighboring nodes, and (ii) contamination and recovery reactions. The dynamics are mapped into a growing process by activating lattice nodes with successful contaminations where activated nodes are interpreted as infection sources. In all simulations, the epidemic starts with one infected individual in a lattice filled with susceptibles. Our results indicate a phase transition in the dynamic percolation universality class controlled by the population size, irrespective of diffusion rates $D_S$ and $D_I$ and a subexponential growth of the epidemics in the percolation threshold.

cond-mat.stat-mech↗

arXiv2024

Critical Short-Time Behavior of Majority-Vote Model on Scale-Free Networks

We discuss the short-time behavior of the majority vote dynamics on scale-free networks at the critical threshold. We introduce a heterogeneous mean-field theory on the critical short-time behavior of the majority-vote model on scale-free networks. In addition, we also compare the heterogeneous mean-field predictions with extensive Monte Carlo simulations of the short-time dependencies of the order parameter and the susceptibility. We obtained a closed expression for the dynamical exponent $z$ and the time correlation exponent $ν_\parallel$. Short-time scaling is compatible with a non-universal critical behavior for $5/2 < γ< 7/2$, and for $γ\geq 7/2$, we have the mean-field Ising criticality with additional logarithmic corrections for $γ=7/2$, in the same way as the stationary scaling.

cond-mat.stat-mech↗

arXiv2024

Estimating thresholds for asynchronous susceptible-infected-removed model on complex networks

We use the pair heterogeneous mean-field (PHMF) approximation for an asynchronous version of the susceptible-infected-removed (SIR) model to estimate the epidemic thresholds on complex quenched networks. Our results indicate an improvement compared to the heuristic heterogeneous mean-field theory developed for one vertex (HMF) when the dynamic evolves on top random regular and power-law networks. However, there is a slight overestimation of the transition point for the later network type. We also analyze scaling for random regular networks near the thresholds. For this region, collapses were shown at the subcritical and supercritical phases.

physics.soc-ph↗

arXiv2023

Is Kaniadakis $κ$-generalized statistical mechanics general?

In this Letter we introduce some field-theoretic approach for computing the critical properties of systems undergoing continuous phase transitions governed by the $κ$-generalized statistics, namely $κ$-generalized statistical field theory. In particular, we show, by computations through analytic and simulation results, that the $κ$-generalized Ising-like systems are not capable of describing the nonconventional critical properties of real imperfect crystals, \emph{e. g.} of manganites, as some alternative generalized theory is, namely nonextensive statistical field theory, as shown recently in literature. Although $κ$-Ising-like systems do not depend on $κ$, we show that a few distinct systems do. Thus the $κ$-generalized statistical field theory is not general, \emph{i. e.} it fails to generalize Ising-like systems for describing the critical behavior of imperfect crystals, and must be discarded as one generalizing statistical mechanics. For the latter systems we present the physical interpretation of the theory by furnishing the general physical interpretation of the deformation $κ$-parameter.

hep-th↗

arXiv2023

Droplet Finite-Size Scaling of the Majority Vote Model on Quenched Scale-Free Networks

We consider the Majority Vote model coupled with scale-free networks. Recent works point to a non-universal behavior of the Majority Vote model, where the critical exponents depend on the connectivity while the network's effective dimension $D_\mathrm{eff}$ is unity for a degree distribution exponent $5/2<γ<7/2$. We present a finite-size theory of the Majority Vote Model for uncorrelated networks and present generalized scaling relations with good agreement with Monte-Carlo simulation results. The presented finite-size theory has two main sources of size dependence. The first source is an external field describing a mass media influence on the consensus formation and the second source is the scale-free network cutoff. The model indeed presents non-universal critical behavior where the critical exponents depend on the degree distribution exponent $5/2<γ<7/2$. For $γ\geq 7/2$, the model is on the same universality class of the Majority Vote model on Erdös-Renyi random graphs, while for $γ=7/2$, the critical behavior presents additional logarithmic corrections.

cond-mat.stat-mech↗

arXiv2023

Machine Learning based tool for CMS RPC currents quality monitoring

The muon system of the CERN Compact Muon Solenoid (CMS) experiment includes more than a thousand Resistive Plate Chambers (RPC). They are gaseous detectors operated in the hostile environment of the CMS underground cavern on the Large Hadron Collider where pp luminosities of up to $2\times 10^{34}$ $\text{cm}^{-2}\text{s}^{-1}$ are routinely achieved. The CMS RPC system performance is constantly monitored and the detector is regularly maintained to ensure stable operation. The main monitorable characteristics are dark current, efficiency for muon detection, noise rate etc. Herein we describe an automated tool for CMS RPC current monitoring which uses Machine Learning techniques. We further elaborate on the dedicated generalized linear model proposed already and add autoencoder models for self-consistent predictions as well as hybrid models to allow for RPC current predictions in a distant future.

physics.ins-det↗

arXiv2022

Epidemic Outbreaks on Quenched Scale-Free Networks

We present a finite-size scaling theory of a contact process with permanent immunity on uncorrelated scale-free networks. We model an epidemic outbreak by an analog of the susceptible-infected-removed model where an infected individual attacks only one susceptible in a time unit in a way we can expect a non-vanishing critical threshold at scale-free networks. As we already know, the susceptible-infected-removed model can be mapped in a bond percolation process, allowing us to compare the critical behavior of site and bond universality classes on networks. We used the external field finite-scale theory, where the dependence on the finite size enters the external field defined as the initial number of infected individuals. We can impose the scale of the external field as $N^{-1}$. The system presents an epidemic-endemic phase transition where the critical behavior obeys the mean-field universality class, as we show theoretically and by simulations.

cond-mat.stat-mech↗

arXiv2022

RPC based tracking system at CERN GIF++ facility

With the HL-LHC upgrade of the LHC machine, an increase of the instantaneous luminosity by a factor of five is expected and the current detection systems need to be validated for such working conditions to ensure stable data taking. At the CERN Gamma Irradiation Facility (GIF++) many muon detectors undergo such studies, but the high gamma background can pose a challenge to the muon trigger system which is exposed to many fake hits from the gamma background. A tracking system using RPCs is implemented to clean the fake hits, taking profit of the high muon efficiency of these chambers. This work will present the tracking system configuration, used detector analysis algorithm and results.

physics.ins-det↗

arXiv2022

Phase Diagram of the Contact Process on Barabasi-Albert Networks

We show results for the contact process on Barabasi networks. The contact process is a model for an epidemic spreading without permanent immunity that has an absorbing state. For finite lattices, the absorbing state is the true stationary state, which leads to the need for simulation of quasi-stationary states, which we did in two ways: reactivation by inserting spontaneous infected individuals, or by the quasi-stationary method, where we store a list of active states to continue the simulation when the system visits the absorbing state. The system presents an absorbing phase transition where the critical behavior obeys the Mean Field exponents $β=1$, $γ'=0$, and $ν=2$. However, the different quasi-stationary states present distinct finite-size logarithmic corrections. We also report the critical thresholds of the model as a linear function of the network connectivity inverse $1/z$, and the extrapolation of the critical threshold function for $z \to \infty$ yields the basic reproduction number $R_0=1$ of the complete graph, as expected. Decreasing the network connectivity leads to the increase of the critical basic reproduction number $R_0$ for this model.

physics.soc-ph↗

arXiv2021

Upgrade of the CMS Resistive Plate Chambers for the High Luminosity LHC

During the upcoming High Luminosity phase of the Large Hadron Collider (HL-LHC), the integrated luminosity of the accelerator will increase to 3000 fb$^{-1}$. The expected experimental conditions in that period in terms of background rates, event pileup, and the probable aging of the current detectors present a challenge for all the existing experiments at the LHC, including the Compact Muon Solenoid (CMS) experiment. To ensure a highly performing muon system for this period, several upgrades of the Resistive Plate Chamber (RPC) system of the CMS are currently being implemented. These include the replacement of the readout system for the present system, and the installation of two new RPC stations with improved chamber and front-end electronics designs. The current overall status of this CMS RPC upgrade project is presented.

physics.ins-det↗

arXiv2021

Modified Epidemic Diffusive Process on the Apollonian Network

We present an analysis of an epidemic spreading process on the Apollonian network that can describe an epidemic spreading in a non-sedentary population. The modified diffusive epidemic process was employed in this analysis in a computational context by means of the Monte Carlo method. Our model has been useful for modeling systems closer to reality consisting of two classes of individuals: susceptible (A) and infected (B). The individuals can diffuse in a network according to constant diffusion rates $D_{A}$ and $D_{B}$, for the classes A and B, respectively, and obeying three diffusive regimes, i.e., $D_{A} D_{B}$. Into the same site $i$, the reaction occurs according to the dynamical rule based on Gillespie's algorithm. Finite-size scaling analysis has shown that our model exhibit continuous phase transition to an absorbing state with a set of critical exponents given by $β/ν=0.66(1)$, $1/ν=0.46(2)$, and $γ/ν=-0.24(2)$ common to every investigated regime. In summary, the continuous phase transition, characterized by this set of critical exponents, does not have the same exponents of the Mean-Field universality class in both regular lattices and complex networks.

cond-mat.stat-mech↗