Hybrid attacks coordinate the exploitation of vulnerabilities across domains to undermine trust in authorities and cause social unrest. Whilst such attacks have primarily been seen in active conflict zones, there is growing concern about the potential harm that can be caused by hybrid attacks more generally and a desire to discover how better to identify and react to them. In addressing such threats, it is important to be able to identify and understand an adversary’s behaviour. Game theory is the approach predominantly used in security and defence literature for this purpose. However, the underlying rationality assumption, the equilibrium concept of game theory, as well as the need to make simplifying assumptions can limit its use in the study of emerging threats. To study hybrid threats, we present a novel agent-based model in which, for the first time, agents use reinforcement learning to inform their decisions. This model allows us to investigate the behavioural strategies of threat agents with hybrid attack capabilities as well as their broader impact on the behaviours and opinions of other agents.

This BNE-informed ABM ultimately aims to provide a more realistic description of complicated pedestrian behaviours especially in high-density and life-threatening situations. Bayesian Nash Equilibrium (BNE) was adopted to reproduce interactive decision-making process among rational and game-playing agents. The implementations of 3 behavioural models, which are Shortest Route (SR) model, Random Follow (RF) model, and BNE model, make it possible to simulate emergent patterns of pedestrian behaviours (e.g. herding and self-organised queuing behaviours, etc.) in emergency situations.

According to the common features of previous mass trampling accidents, a series of simulation experiments were performed in space with 3 types of barriers, which are Horizontal Corridors, Vertical Corridors, and Random Squares, standing for corridors, bottlenecks and intersections respectively, to investigate emergent behaviours of evacuees in varied constricted spatial environments. The output of this ABM has been available at https://data.mendeley.com/datasets/9v4byyvgxh/1.

This ABM aims to introduce a new individual decision-making model, BNE into the ABM of pedestrian evacuation to properly model individual behaviours and motions in emergency situations. Three types of behavioural models has been developed, which are Shortest Route (SR) model, Random Follow (RF) model, and BNE model, to better reproduce evacuation dynamics in a tunnel space. A series of simulation experiments were conducted to evaluate the simulating performance of the proposed ABM.

The simulation experiment is for studying the influence of external supervision services on combating corruption.
Algorithm: evolutionary game theory

This project is based on a Jupyter Notebook that describes the stepwise implementation of the EWA model in bi-matrix ( 2×2 ) strategic-form games for the simulation of economic learning processes. The output is a dataset with the simulated values of Attractions, Experience, selected strategies, and payoffs gained for the desired number of rounds and periods. The notebook also includes exploratory data analysis over the simulated output based on equilibrium, strategy frequencies, and payoffs.

This project combines game theory and genetic algorithms in a simulation model for evolutionary learning and strategic behavior. It is often observed in the real world that strategic scenarios change over time, and deciding agents need to adapt to new information and environmental structures. Yet, game theory models often focus on static games, even for dynamic and temporal analyses. This simulation model introduces a heuristic procedure that enables these changes in strategic scenarios with Genetic Algorithms. Using normalized 2x2 strategic-form games as input, computational agents can interact and make decisions using three pre-defined decision rules: Nash Equilibrium, Hurwicz Rule, and Random. The games then are allowed to change over time as a function of the agent’s behavior through crossover and mutation. As a result, strategic behavior can be modeled in several simulated scenarios, and their impacts and outcomes can be analyzed, potentially transforming conflictual situations into harmony.

The purpose of the model is to collect information on human decision-making in the context of coalition formation games. The model uses a human-in-the-loop approach, and a single human is involved in each trial. All other agents are controlled by the ABMSCORE algorithm (Vernon-Bido and Collins 2020), which is an extension of the algorithm created by Collins and Frydenlund (2018). The glove game, a standard cooperative game, is used as the model scenario.

The intent of the game is to collection information on the human players behavior and how that compares to the computerized agents behavior. The final coalition structure of the game is compared to an ideal output (the core of the games).

ICARUS is a multi-agent compliance inspection model (ICARUS - Inspecting Compliance to mAny RUleS). The model is applicable to environments where an inspection agency, via centrally coordinated inspections, examines compliance in organizations which must comply with multiple provisions (rules). The model (ICARUS) contains 3 types of agents: entities, inspection agency and inspectors / inspections. ICARUS describes a repeated, simultaneous, non-cooperative game of pure competition. Agents have imperfect, incomplete, asymmetric information. Entities in each move (tick) choose a pure strategy (comply/violate) for each rule, depending on their own subjective assessment of the probability of the inspection. The Inspection Agency carries out the given inspection strategy.

A more detailed description of the model is available in the .nlogo file.
Full description of the model (in line with the ODD+D protocol) and the analysis of the model (including verification, validation and sensitivity analysis) can be found in the attached documentation.

This model simulates the Hawk-Dove game as first described by John Maynard Smith, and further elaborated by Richard Dawkins in “The Selfish Gene”. In the game, two strategies, Hawks and Doves, compete against each other, and themselves, for reproductive benefits. A third strategy can be introduced, Retaliators, which act like either Hawks or Doves, depending on the context.

The purpose of the model is to generate coalition structures of different glove games, using a specially designed algorithm. The coalition structures can be are later analyzed by comparing them to core partitions of the game used. Core partitions are coalition structures where no subset of players has an incentive to form a new coalition.

The algorithm used in this model is an advancement of the algorithm found in Collins & Frydenlund (2018). It was used used to generate the results in Vernon-Bido & Collins (2021).