Behavior Of Agents Research Articles

Anisotropic contractile particle model with avoidance for simulating pedestrian navigation in dilute and dense systems

We introduce a novel rule-based pedestrian simulation model that extends an existing one by incorporating enhanced particle contact rules and evasive maneuver capabilities for collision avoidance. The model is calibrated using individual trajectory data from avoidance experiments. With the same set of parameters, the new model satisfactorily replicates experimental observations, including the avoidance behavior of individual agents, fundamental diagrams of unidirectional and bidirectional flows, as well as specific flow rates through bottlenecks.

Improving deep learning with prior knowledge and cognitive models: A survey on enhancing explainability, adversarial robustness and zero-shot learning

We review current and emerging knowledge-informed and brain-inspired cognitive systems for realizing adversarial defenses, eXplainable Artificial Intelligence (XAI), and zero-shot or few-shot learning. Data-driven machine learning models have achieved remarkable performance and demonstrated capabilities surpassing humans in many applications. Yet, their inability to exploit domain knowledge leads to serious performance limitations in practical applications. In particular, deep learning systems are exposed to adversarial attacks, which can trick them into making glaringly incorrect decisions. Moreover, complex data-driven models typically lack interpretability or explainability, i.e., their decisions cannot be understood by human subjects. Furthermore, models are usually trained on standard datasets with a closed-world assumption. Hence, they struggle to generalize to unseen cases during inference in practical open-world environments, thus, raising the zero- or few-shot generalization problem. Although many conventional solutions exist, explicit domain knowledge, brain-inspired neural networks and cognitive architectures offer powerful new dimensions towards alleviating these problems. Prior knowledge is represented in appropriate forms like mathematical relations, logic rules, knowledge graphs, and large language models (LLMs). and incorporated in deep learning frameworks to improve performance. Brain-inspired cognition methods use computational models that mimic the human brain to enhance intelligent behavior in artificial agents and autonomous robots. Ultimately, these models achieve better explainability, higher adversarial robustness and data-efficient learning, and can, in turn, provide insights for cognitive science and neuroscience—that is, to deepen human understanding on how the brain works in general, and how it handles these problems.

Adaptive Learning in Agent-Based Models: An Approach for Analyzing Human Behavior in Pandemic Crowding

This study assesses the impact of incorporating an adaptive learning mechanism into an agent-based model simulating behavior on a university campus during a pandemic outbreak, with the particular case of the COVID-19 pandemic. Our model not only captures individual behavior, but also serves as a powerful tool for assessing the efficacy of geolocalized policies in addressing campus overcrowding and infections. The main objective is to demonstrate RL’s effectiveness in representing agent behavior and optimizing control policies through adaptive decision-making in response to evolving pandemic dynamics. By implementing RL, we identify different temporal patterns of overcrowding violations, shedding light on the complexity of human behavior within semi-enclosed environments. While we successfully reduce campus overcrowding, the study recognizes its limited impact on altering the pandemic’s course, underlining the importance of comprehensive epidemic control strategies. Our research contributes to the understanding of adaptive learning in complex systems and provides insights for shaping future public health policies in similar community settings. It emphasizes the significance of considering individual decision-making influenced by adaptive learning, implementing targeted interventions, and the role of geospatial elements in pandemic control. Future research directions include exploring various parameter settings and updating representations of the disease’s natural history to enhance the applicability of these findings. This study offers valuable insights into managing pandemics in community settings and highlights the need for multifaceted control strategies.

Breakdown characteristics and influencing factors of live 500 kV insulators sprayed by extinguishing agents using helicopters

Regarding the problem that helicopters cannot spray live transmission conductors and wildfires directly under an insulator, we built a full-scale 500 kV insulator flashover test platform to simulate high-altitude helicopters spraying fire extinguishing agents. The chemical formulation, spray intensity, and fragmentation method of the fire extinguishing agents were varied. We simulated the breakdown characteristics of insulators when helicopters spray fire extinguishing agents, revealing the mechanism of high-altitude live fire extinguishment for high-spray-intensity and high-conductivity agents. Furthermore, an insulation performance verification test of a helicopter spraying live equipment at different flight speeds and altitudes was carried out, and the behavior of the fire extinguishing agents was divided into a five-zone diffusion law consisting of the (1) water column, (2) continuous water block, (3) semi-continuous water body, (4) large droplet particles, and (5) small droplet particles. We propose a spraying live transmission line method in which the helicopter flight height and speed jointly control the particle size of the fire extinguishing agent. When the particle size of the fire extinguishing agent at the terminal is controlled to 560–4000 μm, the insulation performance of the fire extinguishing agent can be effectively improved. During high-incidence periods of wildfires, such as the Spring Festival and Qingming Festival in 2023, on-site firefighting on the Hunan power grid was performed using helicopters to spray fire extinguishing agents from top to bottom through live transmission conductors to extinguish wildfire disasters directly below the transmission conductors. Neither transmission line flashovers nor power outages occurred when the fires were extinguished.

Adsorption behavior of VX nerve agent on X12Y12 nanocages: a density functional theory study

Adsorption behavior of VX nerve agent on X12Y12 nanocages: a density functional theory study

An autonomous coverage path planning algorithm for maritime search and rescue of persons-in-water based on deep reinforcement learning

The prevalence of maritime transportation and operations is increasing, leading to a gradual increase in drowning accidents at sea. In the context of maritime search and rescue (SAR), it is essential to develop effective search plans to improve the survival probability of persons-in-water (PIWs). However, conventional SAR search plans typically use predetermined patterns to ensure complete coverage of the search area, disregarding the varying probabilities associated with the PIW distribution. To address this issue, this study has proposed a maritime SAR vessel coverage path planning framework (SARCPPF) suitable for multiple PIWs. This framework comprises three modules, namely, drift trajectory prediction, the establishment of a multilevel search area environment model, and coverage search. First, sea area-scale drift trajectory prediction models were employed using the random particle simulation method to forecast drift trajectories. A hierarchical probability environment map model was established to guide the SAR of multiple SAR units. Subsequently, we integrated deep reinforcement learning with a reward function that encompasses multiple variables to guide the navigation behavior of ship agents. We developed a coverage path planning algorithm aimed at maximizing the success rates within a limited timeframe. The experimental results have demonstrated that our model enables vessel agents to prioritize high-probability regions while avoiding repeated coverage.

Innovation-facilitating networks create inequality.

Theories of innovation often balance contrasting views that either smart people create smart things or smartly constructed institutions create smart things. While population models have shown factors including population size, connectivity and agent behaviour as crucial for innovation, few have taken the individual-central approach seriously by examining the role individuals play within their groups. To explore how network structures influence not only population-level innovation but also performance among individuals, we studied an agent-based model of the Potions Task, a paradigm developed to test how structure affects a group's ability to solve a difficult exploration task. We explore how size, connectivity and rates of information sharing in a network influence innovation and how these have an impact on the emergence of inequality in terms of agent contributions. We find, in line with prior work, that population size has a positive effect on innovation, but also find that large and small populations perform similarly per capita; that many small groups outperform fewer large groups; that random changes to structure have few effects on innovation in the task; and that the highest performing agents tend to occupy more central positions in the network. Moreover, we show that every network factor which improves innovation leads to a proportional increase in inequality of performance in the network, creating 'genius effects' among otherwise 'dumb' agents in both idealized and real-world networks.

3D relative directions based evolutionary computation for UAV-to-UAV interaction in swarm intelligence enabled decentralized networks

Swarm intelligence (SI) is the collective behavior of several intelligent agents (IAs), and unmanned aerial vehicles (UAV) or drones are widely used as IA in SI, of which UAV-based decentralized networks are one. Since the UAV can move in different directions without any surface in the 3D environment, the problem of UAV-to-UAV (UAV2UAV) based interaction is observed in UAV-based SI processes. In a UAV-based network, each UAV needs to follow the 3D flying pattern, and if there is any problem in determining the 3D position pattern, the network loses its arrangement, which may cause chaos in the entire network system. 3D movement of biological animal is observed based on these six relative directions: right, left, up, down, back, and front. The aim of this study is to establish an IA-enabled interaction based on the mentioned 6 relative directions in the UAV-based SI process to overcome the raised issues. First, the article presents various relative directional 3D positions of a UAV and then proposes a novel UAV-to-UAV Relative Directional Interaction (UAV2UAV-RDI) model. Here, two different algorithms are performed on two different UAVs to execute an optimization-based evolutionary computation for detecting and tracking the relative directional 3D position of a UAV. Consequently, according to the findings of this research, the interaction between two UAVs occurs and one UAV can track the relative directional 3D position of the other UAV. The performance of the proposed model is compared with the three models where the proposed model performed more directionally. Additionally, we also point out the limitations of the proposed model for future improvements.

COSMOPOLITAN COMMUNICATION BEHAVIOR AMONG AGRICULTURAL EXTENSION WORKERS IN THE USE OF INFORMATION MEDIA IN THE DIGITAL ERA AT CIANJUR DISTRICT

The digital era has fundamentally brought changes to various agricultural extension activities in carrying out communication activities. As is well known, extension agents as communication managers who serve as development agents are required and actively seeking information. The process seeking extension agent information is related to the media used and the cosmopolitan of the extension agents in dealing with information sources outside the village. This paper that the result of research that aims to describe the cosmopolite communication behavior of agricultural instructors and analyze the factors related to the cosmopolitan communication behavior of agricultural instructors in the digital era. The research location is in Cianjur District North Region, West Java. The research method used was a survey with a purposive sampling method of 5 UPTDs located in the North Region, obtained a sample of 80 extension workers. Data collection used questionnaires as primary data and in-depth interviews with related agencies as secondary data. The score calculation uses by analyzing the data to measure the level of cosmopolite communication behavior of the extension agents and the factors related to the cosmopolitan communication behavior of the extension agents. The results showed that the cosmopolite communication behavior of the extension agents in seeking information outside the village was still moderate. Then, age and years of service negatively correlated with the cosmopolitan communication behavior of the extension agents. It proves that the longer the extension worker works and the older the extension worker is, it does not indicate a cosmopolitan position of the extension agent in seeking information outside the village. Print media and electronic media are still the means of extension agents seeking information through the media, which is indicated by the research results that the accessibility of print and electronic media that positively correlated with the instructor's cosmopolite communication.
 
 Keywords : Communication Behaviour, Agricultural Extension, Digital Era

Reinforcement Learning-Based Formation Pinning and Shape Transformation for Swarms

Swarm models hold significant importance as they provide the collective behavior of self-organized systems. Boids model is a fundamental framework for studying emergent behavior in swarms systems. It addresses problems related to simulating the emergent behavior of autonomous agents, such as alignment, cohesion, and repulsion, to imitate natural flocking movements. However, traditional models of Boids often lack pinning and the adaptability to quickly adapt to the dynamic environment. To address this limitation, we introduce reinforcement learning into the framework of Boids to solve the problem of disorder and the lack of pinning. The aim of this approach is to enable drone swarms to quickly and effectively adapt to dynamic external environments. We propose a method based on the Q-learning network to improve the cohesion and repulsion parameters in the Boids model to achieve continuous obstacle avoidance and maximize spatial coverage in the simulation scenario. Additionally, we introduce a virtual leader to provide pinning and coordination stability, reflecting the leadership and coordination seen in drone swarms. To validate the effectiveness of this method, we demonstrate the model’s capabilities through empirical experiments with drone swarms, and show the practicality of the RL-Boids framework.

Multi-agent reinforcement learning for network routing in integrated access backhaul networks

In this study, we examine the problem of downlink wireless routing in integrated access backhaul (IAB) networks involving fiber-connected base stations, wireless base stations, and multiple users. Physical constraints prevent the use of a central controller, leaving base stations with limited access to real-time network conditions. These networks operate in a time-slotted regime, where base stations monitor network conditions and forward packets accordingly. Our objective is to maximize the arrival ratio of packets, while simultaneously minimizing their latency. To accomplish this, we formulate this problem as a multi-agent partially observed Markov Decision Process (POMDP). Moreover, we develop an algorithm that uses Multi-Agent Reinforcement Learning (MARL) combined with Advantage Actor Critic (A2C) to derive a joint routing policy on a distributed basis. Due to the importance of packet destinations for successful routing decisions, we utilize information about similar destinations as a basis for selecting specific-destination routing decisions. For portraying the similarity between those destinations, we rely on their relational base-station associations, i.e., which base station they are currently connected to. Therefore, the algorithm is referred to as Relational Advantage Actor Critic (Relational A2C). To the best of our knowledge, this is the first work that optimizes routing strategy for IAB networks. Further, we present three types of training paradigms for this algorithm in order to provide flexibility in terms of its performance and throughput. Through numerical experiments with different network scenarios, Relational A2C algorithms were demonstrated to be capable of achieving near-centralized performance even though they operate in a decentralized manner in the network of interest. Based on the results of those experiments, we compare Relational A2C to other reinforcement learning algorithms, like Q-Routing and Hybrid Routing. This comparison illustrates that solving the joint optimization problem increases network efficiency and reduces selfish agent behavior.

A data-driven agent-based simulation of the public bicycle-sharing system in Sejong city

In this study, we propose a data-driven agent-based model for simulating Eoulling, the public bicycle-sharing systems (PBSSs) in Sejong City, the administrative capital of South Korea. Most existing models for PBSSs based on top-down approaches have limitations in reflecting Eoulling users’ behavioral characteristics and analyzing their convenience. Unlike these, the proposed model is based on a bottom-up approach of agent-based simulation. We model each user as an agent to capture their bicycle rental and return behaviors, and analyze user convenience through interactions with bicycle station agent models. To improve model fidelity, multiple parameters for determining agent behaviors are extracted from the actual operations data of Eoulling, along with the population and geographic information of Sejong City. The validation results showed that the proposed model accurately describes the behavioral characteristics. We provide a workable solution addressing multiple concerns with Eoulling by evaluating its utilization and user convenience in virtual scenarios via model simulations.

Hope and life satisfaction in Palestinian children victim of military violence: The predictive role of agency, potentially traumatic experiences and symptoms of trauma

BackgroundChildren exposed political violence deploy resources to maintain functioning, hope and life satisfaction. ObjectiveWe sought to explore whether or not children promote hope and life satisfaction trough agency, psychological difficulties, potentially traumatic experiences and symptoms in Palestine. Participants and setting965 children (494 males and 471 females) in multiple geographical contexts, and areas were involved. MethodsWe administered the War Child Agency Assessment Scale, Child Hope Scale, Multilevel Students'Life Satisfaction Scale-Bref, the Strength and difficulties scale, the Child Revised Impact of events Scale, and Trauma Checklist, and performed regression analysis; hope and life satisfaction were dependent and agency, strength and difficulties, trauma symptoms and traumatic events independent variables. ResultsSpecific forms of agency predicted life satisfaction (β = 0.219; ** p < .01, social agency; β = 0.11; ** p < .01, with agency in education) and hope (β = 0.07; ** p < .05, agency on free movement), while mental difficulties (conduct problems, β = −0.09; ** p < .01; hyperactivity, β = −0.07; ** p < .05; β = −0.15; ** p < .01 with life satisfaction) (conduct problems, β = −0.06; ** p < .05, and difficulties in pro-social behaviour, β = −0.21; ** p < .01 with hope), traumatic events (β = −0.16; ** p < .01, with life satisfaction; β = −0.15; ** p < .01, with hope) and trauma symptoms (β = −0.09; ** p < .05, with hope) were negatively associated with the dependents variables. ConclusionsWe found a positive role of social, educational, and freedom of movement agentic behaviours in fostering hope and life satisfaction.

Distributed Coordination of Charging Stations With Shared Energy Storage in a Distribution Network

Electric vehicle (EV) charging stations have experienced rapid growth, whose impacts on the power grid have become non-negligible. Though charging stations can install energy storage to reduce their impacts on the grid, the conventional “one charging station, one energy storage” method may be uneconomical due to the high upfront cost of energy storage. Shared energy storage can be a potential solution. However, effective management of charging stations with shared energy storage in a distribution network is challenging due to the complex coupling, competing interests, and information asymmetry between different agents. To address the aforementioned challenges, this paper first proposes an equilibrium model to characterize the interaction among charging stations, shared energy storage, and the distribution network. We prove that the equilibrium coincides with the centralized optimization result with trading prices equaling the value of dual variables at optimum. Then, to achieve the efficient equilibrium, a distributed coordination mechanism with a prediction and a correction step is developed to guide the behaviors of different agents with proof of convergence. Numerical experiments and comprehensive performance comparisons are conducted to validate the theoretical results and show the advantages of the proposed mechanism.

Future Bias and Regret

Although the rationality of future bias figures crucially in various metaphysical and ethical arguments (Prior 1959; Parfit 1984; Fischer 2020), many philosophers have challenged future bias as being either arbitrarily motivated or irrational (Dougherty 2011; Suhler and Callender 2012; Greene and Sullivan 2015). In particular, Greene and Sullivan (2015) have claimed that future bias is irrational because it implicates two kinds of irrational planning behaviors in agents who seek to avoid regret. In this paper, I join others (Dorsey 2016; Tarsney 2017) in arguing against their claims, but for different reasons that highlight the relationship between the alleged irrational planning behaviors and certain features of regret that it shares with future bias. First, regret is dynamic, involving preferences that change over time and in inconsistent ways. Second, regret comes in degrees, meaning that we can rank our potential regrets. Because regret has these features, I explain why the future-biased agents in Greene and Sullivan’s cases do not need to act in irrational ways to avoid regret.

A hierarchical multi-agent allocation-action learning framework for multi-subtask games

Great progress has been made in the domain of multi-agent reinforcement learning in recent years. Most work concentrates on solving a single task by learning the cooperative behaviors of agents. However, many real-world problems are normally composed of a set of subtasks in which the execution order follows a certain procedure. Cooperative behaviors should be learned on the premise that agents are first allocated to those subtasks. In this paper, we propose a hierarchical framework for learning the dynamic allocation of agents among subtasks, as well as cooperative behaviors. We design networks corresponding to agents and subnetworks, respectively, which constitute the whole hierarchical network. For the upper layer, a novel allocation learning mechanism is devised to map an agent network to a subtask network. Each agent network could only be assigned to only one subtask network at each time step. For the lower layer, an action learning module is designed to compute appropriate actions for each agent with the allocation result. The agent networks together with the subtask networks are updated by a common reward obtained from the environment. To evaluate the effectiveness of our framework, we conduct experiments in two challenging environments, i.e., Google Research Football and SAVETHECITY. Empirical results show that our framework achieves much better performance than other recent methods.

Determinants of Indonesian Gen Z’s purchase behavior on online travel platforms: Extending UTAUT model

Understanding Gen Z’s purchase behavior in online travel agents is essential to effectively engage and meet the unique preferences of this generation, fostering long-term loyalty and satisfaction. Utilizing the unified theory of acceptance and use of technology (UTAUT) as the theoretical foundation, this study aims to analyze the impact of performance expectancy, effort expectancy, social influence, facilitating condition, and trust on purchase decision of flight tickets through online travel agent platforms. The data were collected through an online survey of 253 Gen Z users of online travel agent applications in Indonesia, such as Traveloka, Tiket.com, Pegipegi.com, Agoda, and Booking.com. The study employed PLS-SEM to test the hypotheses. The results indicate that performance expectancy, effort expectancy, social influence, and facilitating conditions influence trust (t-value 1.645, p-value &amp;amp;lt; 0.05). Further, performance expectancy, effort expectancy, and facilitating conditions influence purchase decisions (t-value 1.645, p-value &amp;amp;lt; 0.05). However, social influence does not significantly affect purchase decisions (t-value 1.041, p-value &amp;amp;gt; 0.05). The analysis also shows that trust fully mediates the relationship between social influence and purchase decisions, while no mediating effect is identified in the relationship between effort expectancy and purchase decisions. By investigating the key factors contributing to Gen Z’s buying behavior in online travel agent platforms, this paper provides valuable insights for online travel businesses to effectively engage and cater to Gen Z’s unique needs and preferences.

Crowd model calibration at strategic, tactical, and operational levels: Full-spectrum sensitivity analyses show bottleneck parameters are most critical, followed by exit-choice-changing parameters

ABSTRACT Crowd motion simulation requires specification of a range of parameters, each reflecting certain aspects of agent behavior. But what parameters matter the most? Are they all equally important? The question is important given that available data and resources for parameter calibration are limited, and priorities often need to be made. Here, for the first time, a full-spectrum sensitivity analysis of crowd model parameters is reported. It is shown that estimates of simulated evacuation time are, by far, most dependent on the value of locomotion/operational parameters, especially those that determine discharge rate at bottlenecks. The next most critical set of parameters are those that influence change of direction choices. If a crowd simulation model fails to reproduce bottleneck flows accurately, efforts to refine other modeling layers will be in vain. Similarly, if the model fails to represent exit choice adaptation/changing accurately, efforts to refine the exit choice model will be fruitless.

Unique portal venous phase imaging discordance between CEUS and MRI: a valuable predictor of intrahepatic cholangiocarcinoma?

We have long noted unique portal venous phase (PVP) imaging discordance of focal liver masses between CEUS, showing rapid marked washout, and MRI, showing progressive or sustained enhancement. We postulate association of this unique discordance with intrahepatic cholangiocarcinoma (ICC) and causal relationship to different contrast agent behavior. We investigate this unique discordance, propose its clinical significance for ICC diagnosis, and confirm further histologic associations. Cases were collected within our CEUS department and from pathology records over a ten-year interval. This retrospective review includes 99 patients, 73 with confirmed ICC and 26 other diagnoses,showing unique PVP discordance. The CEUS and MRI enhancement characteristics were compared for all patients. Unique discordance is identified in 67/73 (92%) ICC and difference between the PVP appearance on MRI and CEUS is statistically significant (p< 0.0001). Arterial phase enhancement did not show statistically significant difference between CEUS and MRI, p> 0.05. Other diagnoses showing unique discordance include especially lymphoma (n=7), sclerosed hemangioma (n=6), HCC (n=4), metastases (n=2), and other rare entities. ICC shows this discrepant intermodality enhancement pattern in a statistically significant number of cases and should be considered along with other LR-M features in at-risk patients. Discordance is also rarely seen in a number of other liver lesions.

Learning of Generalizable and Interpretable Knowledge in Grid-Based Reinforcement Learning Environments

Understanding the interactions of agents trained with deep reinforcement learning is crucial for deploying agents in games or the real world. In the former, unreasonable actions confuse players. In the latter, that effect is even more significant, as unexpected behavior cause accidents with potentially grave and long-lasting consequences for the involved individuals. In this work, we propose using program synthesis to imitate reinforcement learning policies after seeing a trajectory of the action sequence. Programs have the advantage that they are inherently interpretable and verifiable for correctness. We adapt the state-of-the-art program synthesis system DreamCoder for learning concepts in grid-based environments, specifically, a navigation task and two miniature versions of Atari games, Space Invaders and Asterix. By inspecting the generated libraries, we can make inferences about the concepts the black-box agent has learned and better understand the agent's behavior. We achieve the same by visualizing the agent's decision-making process for the imitated sequences. We evaluate our approach with different types of program synthesizers based on a search-only method, a neural-guided search, and a language model fine-tuned on code.