The wisdom of the crowd refers to the phenomenon in which a group of individuals, each making independent decisions, can collectively arrive at highly accurate solutions—often more accurate than any individual within the group. This principle relies heavily on independence: if individual opinions are unbiased and uncorrelated, their errors tend to cancel out when averaged, reducing overall bias. However, in real-world social networks, individuals are often influenced by their neighbors, introducing correlations between decisions. Such social influence can amplify biases, disrupting the benefits of independent voting. This trade-off between independence and interdependence has striking parallels to ensemble learning methods in machine learning. Bagging (bootstrap aggregating) improves classification performance by combining independently trained weak learners, reducing bias. Boosting, on the other hand, explicitly introduces sequential dependence among learners, where each learner focuses on correcting the errors of its predecessors. This process can reinforce biases present in the data even if it reduces variance. Here, we introduce a new meta-algorithm, casting, which captures this biological and computational trade-off. Casting forms partially connected groups (“castes”) of weak learners that are internally linked through boosting, while the castes themselves remain independent and are aggregated using bagging. This creates a continuum between full independence (i.e., bagging) and full dependence (i.e., boosting). This method allows for the testing of model capabilities across values of the hyperparameter which controls connectedness. We specifically investigate classification tasks, but the method can be used for regression tasks as well. Ultimately, casting can provide insights for how real systems contend with classification problems.

This Agent-Based Model is designed to simulate how similarity-based partner selection (homophily) shapes the formation of co-offending networks and the diffusion of skills within those networks. Its purpose is to isolate and test the effects of offenders’ preference for similar partners on network structure and information flow, under controlled conditions.

In the model, offenders are represented as agents with an individual attribute and a set of skills. At each time step, agents attempt to select partners based on similarity preference. When two agents mutually select each other, they commit a co-offense, forming a tie and exchanging a skill. The model tracks the evolution of network properties (e.g., density, clustering, and tie strength) as well as the spread of skills over time.

This simple and theoretical model does not aim to produce precise empirical predictions but rather to generate insights and test hypotheses about the trade-off between network stability and information diffusion. It provides a flexible framework for exploring how changes in partner selection preferences may lead to differences in criminal network dynamics. Although the model was developed to simulate offenders’ interactions, in principle, it could be applied to other social processes involving social learning and skills exchange.

…

This is a computational model to articulate the theory and test some assumption and axioms for the trust model and its relationship to SBH.

Transhumants move their herds based on strategies simultaneously considering several environmental and socio-economic factors. There is no agreement on the influence of each factor in these strategies. In addition, there is a discussion about the social aspect of transhumance and how to manage pastoral space. In this context, agent-based modeling can analyze herd movements according to the strategy based on factors favored by the transhumant. This article presents a reductionist agent-based model that simulates herd movements based on a single factor. Model simulations based on algorithms to formalize the behavioral dynamics of transhumants through their strategies. The model results establish that vegetation, water outlets and the socio-economic network of transhumants have a significant temporal impact on transhumance. Water outlets and the socio-economic network have a significant spatial impact. The significant impact of the socio-economic factor demonstrates the social dimension of Sahelian transhumance. Veterinarians and markets have an insignificant spatio-temporal impact. To manage pastoral space, water outlets should be at least 15 km
from each other. The construction of veterinary centers, markets and the securitization of transhumance should be carried out close to villages and rangelands.

This model simulates the heterogeneity of preferences in a PG game and how the interaction between them affects the dynamics of voluntary contributions. Model is based on the results of a human-based experiment.

EMMIT is an end-user developed agent-based simulation of malaria transmission. The simulation’s development is a case study demonstrating an approach for non-technical investigators to easily develop useful simulations of complex public health problems. We focused on malaria transmission, a major global public health problem, and insecticide resistance (IR), a major problem affecting malaria control. Insecticides are used to reduce transmission of malaria caused by the Plasmodium parasite that is spread by the Anopheles mosquito. However, the emergence and spread of IR in a mosquito population can diminish the insecticide’s effectiveness. IR results from mutations that produce behavioral changes or biochemical changes (such as detoxification enhancement, target site alterations) in the mosquito population that provide resistance to the insecticide. Evolutionary selection for the IR traits reduces the effectiveness of an insecticide favoring the resistant mosquito population. It has been suggested that biopesticides, and specifically those that are Late Life Acting (LLA), could address this problem. LLA insecticides exploit Plasmodium’s approximate 10-day extrinsic incubation period in the mosquito vector, a delay that limits malaria transmission to older infected mosquitoes. Since the proposed LLA insecticide delays mosquito death until after the exposed mosquito has a chance to produce several broods of offspring, reducing the selective pressure for resistance, it delays IR development and gives the insecticide longer effectivity. Such insecticides are designed to slow the evolution of IR thus maintaining their effectiveness for malaria control. For the IR problem, EMMIT shows that an LLA insecticide could work as intended, but its operational characteristics are critical, primarily the mean-time-to-death after exposure and the associated standard deviation. We also demonstrate the simulation’s extensibility to other malaria control measures, including larval source control and policies to mitigate the spread of IR. The simulation was developed using NetLogo as a case study of a simple but useful approach to public health research.

An agent-based model of species interaction on fragmented landscape is developed to address the question, how do population levels of predators and prey react with respect to changes in the patch connectivity as well as changes in the sharpness of threshold dispersal?

A letter sending model with historically informed initial positions to reconstruct communication and archiving processes in the Republic of Letters, the 15th to 17th century form of scholarship.

The model is aimed at historians, willing to formalize historical assumptions about the letter sending process itself and allows in principle to set heterogeneous social roles, e.g. to evaluate the role of gender or social status in the formation of letter exchange networks. The model furthermore includes a pruning process to simulate the loss of letters to critically asses the role of biases e.g. in relation to gender, geographical regions, or power structures, in the creation of empirical letter archives.

Each agent has an initial random topic vector, expressed as a RGB value. The initial positions of the agents are based on a weighted random draw based on data from [2]. In each step, agents generate two neighbourhoods for sending letters and potential targets to move towards. The probability to send letters is a self-reinforcing process. After each sending the internal topic of the receiver is updated as a movement in abstract space by a random amount towards the letters topic.

…

Agent based simulation model for the study of the genetic evolution of sexual recombination and social behavior

Must tax-benefit policy making be limited to the ‘experts’?