CyberSec Research

We consider the scaling behaviour of the fluctuation susceptibility associated with the average activation in the Greenberg-Hastings neural network model and its relation to microscopic spontaneous activation. We found that, as the spontaneous activation probability tends to zero, a clear finite size scaling behaviour in the susceptibility emerges, characterized by critical exponents which follow already known scaling laws. This shows that the spontaneous activation probability plays the role of an external field conjugated to the order parameter of the dynamical activation transition. The roles of different kinds of activation mechanisms around the different dynamical phase transitions exhibited by the model are characterized numerically and using a mean field approximation.

We present a strong theoretical foundation that frames a well-defined family of outer-totalistic network automaton models as a topological generalisation of binary outer-totalistic cellular automata, of which the Game of Life is one notable particular case. These "Life-like network automata" are quantitatively described by expressing their genotype (the mean field curve and Derrida curve) and phenotype (the evolution of the state and defect averages). After demonstrating that the genotype and phenotype are correlated, we illustrate the utility of these essential metrics by tackling the firing squad synchronisation problem in a bottom-up fashion, with results that exceed a 90% success rate.

We investigate the computational complexity of the timed prediction problem in two-dimensional sandpile models. This question refines the classical prediction problem, which asks whether a cell q will eventually become unstable after adding a grain at cell p from a given configuration. The prediction problem has been shown to be P-complete in several settings, including for subsets of the Moore neighborhood, but its complexity for the von Neumann neighborhood remains open. In a previous work, we provided a complete characterization of crossover gates (a key to the implementation of non-planar monotone circuits) for these small neighborhoods, leading to P-completeness proofs with only 4 and 5 neighbors among the eight adjancent cells. In this paper, we introduce the timed setting, where the goal is to determine whether cell q becomes unstable exactly at time t. We distinguish several cases: some neighborhoods support complete timed toolkits (including timed crossover gates) and exhibit P-completeness; others admit timed crossovers but suffer from synchronization issues; planar neighborhoods provably do not admit any timed crossover; and finally, for some remaining neighborhoods, we conjecture that no timed crossover is possible.

We explore how strategic leaps alter the classic rock-paper-scissors dynamics in spatially structured populations. In our model, individuals can expend energy reserves to jump toward regions with a high density of individuals of the species they dominate in the spatial game. This enables them to eliminate the target organisms and gain new territory, promoting species proliferation. Through stochastic, lattice-based simulations, we show that even when the energy allocated to jumping, as opposed to random walking, is low, there is a significant shift in the cyclic dominance balance. This arises from the increased likelihood of the leaping species successfully acquiring territory. Due to the cyclical nature of the game, the dominant species becomes the one that is superior to the jumping species. We investigate how spatial patterns are affected and calculate the changes in characteristic length scales. Additionally, we quantify how saltatory targeting reshapes spatial correlations and drives shifts in population dominance. Finally, we estimate the coexistence probability and find evidence that this behavioural strategy may promote biodiversity among low-mobility organisms but jeopardise long-term coexistence in the case of high-mobility dispersal. These results underscore the profound impact of novel foraging tactics on community structure and provide concrete parameters for ecologists seeking to incorporate behavioural innovation into ecosystem models.

Boolean networks, inspired by gene regulatory networks, were developed to understand the complex behaviors observed in biological systems, with network attractors corresponding to biological phenotypes or cell types. In this article, we present a proof for a conjecture by Williadsen, Triesch and Wiles about upper bounds for the stability of basins of attraction in Boolean networks. We further extend this result from a single basin of attraction to the entire network. Specifically, we demonstrate that the asymptotic upper bound for the robustness and the basin entropy of a Boolean network are negatively linearly related.

Central to the artificial life endeavour is the creation of artificial systems spontaneously generating properties found in the living world such as autopoiesis, self-replication, evolution and open-endedness. While numerous models and paradigms have been proposed, cellular automata (CA) have taken a very important place in the field notably as they enable the study of phenomenons like self-reproduction and autopoiesis. Continuous CA like Lenia have been showed to produce life-like patterns reminiscent, on an aesthetic and ontological point of view, of biological organisms we call creatures. We propose in this paper Flow-Lenia, a mass conservative extension of Lenia. We present experiments demonstrating its effectiveness in generating spatially-localized patters (SLPs) with complex behaviors and show that the update rule parameters can be optimized to generate complex creatures showing behaviors of interest. Furthermore, we show that Flow-Lenia allows us to embed the parameters of the model, defining the properties of the emerging patterns, within its own dynamics thus allowing for multispecies simulations. By using the evolutionary activity framework as well as other metrics, we shed light on the emergent evolutionary dynamics taking place in this system.

A subset of 10 of the 256 elementary cellular automata (ECA) are implemented as a hash function using an error minimization lossy compression algorithm operating on wrapped 4x4 neighborhood cells. All 256 rules are processed and 10 rules in two subsets of 8 are found to have properties that include both error minimization and maximization, unique solutions, a lossy inverse, efficient retroactive hashing, and an application to edge detection. The algorithm parallels the nested powers-of-two structure of the Fast Fourier Transform and Fast Walsh-Hadamard Transform, is implemented in Java, and is built to hash any 2 byte RGB code bitmap.

We present a local offline decoder for topological codes that operates according to a parallelized message-passing framework. The decoder works by passing messages between anyons, with the contents of received messages used to move nearby anyons towards one another. We prove the existence of a threshold, and show that in a system of linear size $L$, decoding terminates with an $O((\log L)^\eta)$ average-case runtime, where $\eta$ is a small constant. For i.i.d Pauli noise, our decoder has $\eta=1$ and a threshold at a noise strength of $p_c\approx 7.3\%$.

Tuning the interface properties of multiphase models is of paramount importance to the final goal of achieving a one-to-one matching with nucleation and cavitation experiments. The surface tension, at the leading order, and the Tolman length, at higher order, play a crucial role in the estimation of the free-energy barrier determining the experimentally observed nucleation rates. The lattice Boltzmann method allows for a computationally efficient modelling approach of multiphase flows, however, tuning results are concerned with the surface tension and neglect the Tolman length. We present a novel perspective that leverages all the degrees of freedom hidden in the forcing stencil of the Shan-Chen multiphase model. By means of the lattice pressure tensor we determine and tune the coefficients of higher-order derivative terms related to surface tension and Tolman length at constant interface width and density ratio. We test the method by means of both hydrostatic and dynamic simulations and demonstrate the dependence of homogeneous nucleation rates on the value of the Tolman length. This work provides a new tool that can be integrated with previously existing strategies thus marking a step forwards to a high-fidelity modelling of phase-changing fluid dynamics.

In this paper, we study the effect of (a)synchronism on the dynamics of elementary cellular automata. Within the framework of our study, we choose five distinct update schemes, selected from the family of periodic update modes: parallel, sequential, block-sequential, block-parallel, and local clocks. Our main measure of complexity is the maximum period of the limit cycles in the dynamics of each rule. In this context, we present a classification of the ECA rule landscape. We classified most elementary rules into three distinct regimes: constant, linear, and superpolynomial. Surprisingly, while some rules exhibit more complex behavior under a broader class of update schemes, others show similar behavior across all the considered update schemes. Although we are able to derive upper and lower bounds for the maximum period of the limit cycles in most cases, the analysis of some rules remains open. To complement the study of the 88 elementary rules, we introduce a numerical simulation framework based on two main measurements: the energy and density of the configurations. In this context, we observe that some rules exhibit significant variability depending on the update scheme, while others remain stable, confirming what was observed as a result of the classification obtained in the theoretical analysis.

A Cellular Automata (CA) rule is presented that can generate "loop patterns" in a 2D grid under fixed boundary conditions. A loop is a cyclically closed path represented by one-cells enclosed by zero-cells. A loop pattern can contain several loops that are not allowed to touch each other. The problem is solved by designing an appropriate set of tiles that can overlap and which are used in the CA rule. Templates are derived from the tiles which are used for local pattern matching. In order to drive the evolution to the desired patterns, noise is injected if the templates do not match or other constraints are not fulfilled. The general CA rule can be specialized by enabling certain conditions, and the characteristics of five rule variants are explained. Simulations illustrate that the CA rule can securely evolve stable loop patterns. The preliminary theoretical analysis of the obtained loop patterns raises many interesting research problems for the future -- several of them have been briefly discussed.

We explore the dynamics of a one-dimensional lattice of state machines on two states and two symbols sequentially updated via a process of "reflexive composition." The space of 256 machines exhibits a variety of behavior, including substitution, reversible "billiard ball" dynamics, and fractal nesting. We show that one machine generates the Sierpinski Triangle and, for a subset of boundary conditions, is isomorphic to cellular automata Rule 90 in Wolfram's naming scheme. More surprisingly, two other machines follow trajectories that map to Rule 90 in reverse. Whereas previous techniques have been developed to uncover preimages of Rule 90, this is the first study to produce such inverse dynamics naturally from the formalism itself. We argue that the system's symmetric treatment of state and message underlies its expressive power.

This project explores speculative evolution through a 3D implementation of Conway's Game of Life, using procedural simulation to generate unfamiliar extraterrestrial organic forms. By applying a volumetric optimized workflow, the raw cellular structures are smoothed into unified, bone-like geometries that resemble hypothetical non-terrestrial morphologies. The resulting forms, strange yet organic, are 3D printed as fossil-like artifacts, presenting a tangible representation of generative structures. This process situates the work at the intersection of artificial life, evolutionary modeling, and digital fabrication, illustrating how simple rules can simulate complex biological emergence and challenge conventional notions of organic form.

The magnetic metamaterials known as Artificial Spin Ice (ASI) are promising candidates for neuromorphic computing, composed of vast numbers of interacting nanomagnets arranged in the plane. Every computing device requires the ability to transform, transmit and store information. While ASI excel at data transformation, reliable transmission and storage has proven difficult to achieve. Here, we take inspiration from the Cellular Automaton (CA), an abstract computing model reminiscent of ASI. In CAs, information transmission and storage can be realised by the ``glider'', a simple structure capable of propagating while maintaining its form. Employing an evolutionary algorithm, we search for gliders in pinwheel ASI and present the simplest glider discovered: the ``snake''. Driven by a global field protocol, the snake moves strictly in one direction, determined by its orientation. We demonstrate the snake, both in simulation and experimentally, and analyse the mechanism behind its motion. The snake provides a means of manipulating a magnetic texture in an ASI with resolution on the order of 100 nm, which could in turn be utilised to precisely control other magnetic phenomena. The integration of data transmission, storage and modification into the same magnetic substrate unlocks the potential for ultra-low power computing devices.

While the lattice Boltzmann method (LBM) has proven robust in areas like general fluid dynamics, heat transfer, and multiphase modeling, its application to mass transfer has been limited. Current modeling strategies often oversimplify the complexities required for accurate and realistic mass transfer simulations in multicomponent miscible mixtures involving external forces. We propose a forcing approach within the explicit velocity-difference LBM framework to address these limitations. Our approach recovers the macroscopic mass conservation equations, the Navier-Stokes equation with external forcing term, and the full Maxwell-Stefan equation for ideal mixtures at low Knudsen numbers. A novel boundary scheme for impermeable solid walls is also suggested to ensure proper mass conservation while effectively managing the spatial interpolations required for multicomponent mixtures with varying molecular masses. We demonstrated the physical consistency and accuracy of the proposed forcing approach through simulations of the ultracentrifuge separation of uranium isotopes and the Loschmidt tube with gravitational effects. Our approach encompasses advanced modeling of species dynamics influenced by force fields, such as those encountered in geological CO$_2$ sequestration in aquifers and oil reservoirs under gravitational fields.

The cornerstones of the Cellular Automaton Interpretation of Quantum Mechanics are its underlying ontological states that evolve by permutations. They do not create would-be quantum mechanical superposition states. We review this with a classical automaton consisting of an Ising spin chain which is then related to the Weyl equation in the continuum limit. Based on this and generalizing, we construct a new ``Necklace of Necklaces'' automaton with a torus-like topology that lends itself to represent the Dirac equation in 1 + 1 dimensions. Special attention has to be paid to its mass term, which necessitates this enlarged structure and a particular scattering operator contributing to the step-wise updates of the automaton. As discussed earlier, such deterministic models of discrete spins or bits unavoidably become quantum mechanical, when only slightly deformed.

We study stationary fluctuations of conserved slow modes in a two-lane model of hardcore particles which are expected to show universal behaviour. Specifically, we focus on the properties of fluctuations at a special umbilic point where the characteristic velocities coincide. At large space and time scales, fluctuations are described by a system of stochastic Burgers equations studied recently in [13]. Our data suggest coupling-dependent scaling functions and, even more surprisingly, coupling-dependent dynamical scaling exponents, distinct from KPZ scaling exponent typical for surface growth processes.

We consider a one-dimensional array of particles interacting via an infinite well potential. We explore the properties of energy spreading from an initial state where only a group of particles has non-zero velocities while others are resting. We characterize anomalous diffusion of the active domain via moments and entropies of the energy distribution. Only in the special cases of a single-well potential (hard-particle gas) and of the distance between the particles being half of the potential width does the diffusion have a single scale; otherwise, a multiscale anomalous diffusion is observed.