Suchergebnisse

Learning in a multi-agent context is a difficult task, especially since the communication with other agents require additional information to be stored and used in the learning process. In this paper, we propose a model of agent memory for command and control in a multi-agent context. We first show why the existing memory models are not sufficient to support automated learning of command and control in a multi-agent context. Then, we describe our model for storing messages and answers. Finally we show on an example that this model allows faster learning and better convergence. ; L'apprentissage de comportement dans un contexte multi-agent est un problème difficile, en particulier parce que la prise en compte de la communication avec les autres agents requiert la mémorisation d'informations spécifiques. Dans cet article, nous présentons les modèles de mémoire existants et nous montrons qu'ils ne permettent pas de gérer l'apprentis-sage de communication. Nous proposons en-suite un modèle de mémoire pour l'apprentis-sage par renforcement des messages de commande (request) et de contrôle (query). Ce modèle permet de gérer l'asynchronisme du système et les attentes de réponses aux messages. Enfin, nous présentons une évaluation de ce modèle sur un exemple simple et nous montrons qu'il construit une politique en un temps raison-nable et avec un espace mémoire réduit.

Learning in a multi-agent context is a difficult task, especially since the communication with other agents require additional information to be stored and used in the learning process. In this paper, we propose a model of agent memory for command and control in a multi-agent context. We first show why the existing memory models are not sufficient to support automated learning of command and control in a multi-agent context. Then, we describe our model for storing messages and answers. Finally we show on an example that this model allows faster learning and better convergence. ; L'apprentissage de comportement dans un contexte multi-agent est un problème difficile, en particulier parce que la prise en compte de la communication avec les autres agents requiert la mémorisation d'informations spécifiques. Dans cet article, nous présentons les modèles de mémoire existants et nous montrons qu'ils ne permettent pas de gérer l'apprentis-sage de communication. Nous proposons en-suite un modèle de mémoire pour l'apprentis-sage par renforcement des messages de commande (request) et de contrôle (query). Ce modèle permet de gérer l'asynchronisme du système et les attentes de réponses aux messages. Enfin, nous présentons une évaluation de ce modèle sur un exemple simple et nous montrons qu'il construit une politique en un temps raison-nable et avec un espace mémoire réduit.

International audience ; Attitude is a key concept in social psychology. The paper presents a novel agent-based model to simulate attitude formation by combining a rational and an emotional components based on cognitive, psychological and social theories. Individuals of the artificial population perceive actions taken by actors such as government or brands, they form an attitude toward them and also communicate the events through a social network. The model outputs are first studied through a functional analysis in which some unique macroscopic behaviors have emerged such as the impact of social groups, the resistance of the population toward disinfor-mation campaigns or the social pressure. We then applied our model on a real world scenario depicting the eeort of French Forces in their stabilization operations in Kapisa (Afghanistan) between and. We calibrated the model parameters based on this scenario and the results of opinion polls that were conducted in the area during the same period about the sentiment of the population toward the Forces. Our model was able to reproduce polls results with a global error under %. Based on these results, we show the diierent dynamics tendencies that emerged among the population by applying a non-supervised classification algorithm.

International audience ; Attitude is a key concept in social psychology. The paper presents a novel agent-based model to simulate attitude formation by combining a rational and an emotional components based on cognitive, psychological and social theories. Individuals of the artificial population perceive actions taken by actors such as government or brands, they form an attitude toward them and also communicate the events through a social network. The model outputs are first studied through a functional analysis in which some unique macroscopic behaviors have emerged such as the impact of social groups, the resistance of the population toward disinfor-mation campaigns or the social pressure. We then applied our model on a real world scenario depicting the eeort of French Forces in their stabilization operations in Kapisa (Afghanistan) between and. We calibrated the model parameters based on this scenario and the results of opinion polls that were conducted in the area during the same period about the sentiment of the population toward the Forces. Our model was able to reproduce polls results with a global error under %. Based on these results, we show the diierent dynamics tendencies that emerged among the population by applying a non-supervised classification algorithm.

International audience ; Attitude is a key concept in social psychology. The paper presents a novel agent-based model to simulate attitude formation by combining a rational and an emotional components based on cognitive, psychological and social theories. Individuals of the artificial population perceive actions taken by actors such as government or brands, they form an attitude toward them and also communicate the events through a social network. The model outputs are first studied through a functional analysis in which some unique macroscopic behaviors have emerged such as the impact of social groups, the resistance of the population toward disinfor-mation campaigns or the social pressure. We then applied our model on a real world scenario depicting the eeort of French Forces in their stabilization operations in Kapisa (Afghanistan) between and. We calibrated the model parameters based on this scenario and the results of opinion polls that were conducted in the area during the same period about the sentiment of the population toward the Forces. Our model was able to reproduce polls results with a global error under %. Based on these results, we show the diierent dynamics tendencies that emerged among the population by applying a non-supervised classification algorithm.

International audience ; This paper presents a multi-agent model for simulating attitude formation and change based on perception and communication in the context of stabilization operations. The originality of our model comes from (1) attitude computation that evaluates information as part of a history relative to the individual (2) a notion of co-responsibility for attitude attribution. We present a military scenario of French operations in Afghanistan along with polls results about the opinion of citizen toward present Forces. Based on these field data, we calibrate the model and show the resulting attitude dynamics. We study the sensibility of the model to the co-responsibility factor.

International audience ; This paper presents a multi-agent model for simulating attitude formation and change based on perception and communication in the context of stabilization operations. The originality of our model comes from (1) attitude computation that evaluates information as part of a history relative to the individual (2) a notion of co-responsibility for attitude attribution. We present a military scenario of French operations in Afghanistan along with polls results about the opinion of citizen toward present Forces. Based on these field data, we calibrate the model and show the resulting attitude dynamics. We study the sensibility of the model to the co-responsibility factor.

International audience ; This paper presents a multi-agent model for simulating attitude formation and change based on perception and communication in the context of stabilization operations. The originality of our model comes from (1) attitude computation that evaluates information as part of a history relative to the individual (2) a notion of co-responsibility forattitude attribution. We present a military scenario of French operations in Afghanistan along with polls results about the opinion of citizen toward present Forces. Based on these eld data, we calibrate the model and show the resulting attitude dynamics. We study the sensibility of the model to the co-responsibility factor.

International audience ; This paper presents a multi-agent model for simulating attitude formation and change based on perception and communication in the context of stabilization operations. The originality of our model comes from (1) attitude computation that evaluates information as part of a history relative to the individual (2) a notion of co-responsibility for attitude attribution. We present a military scenario of French operations in Afghanistan along with polls results about the opinion of citizen toward present Forces. Based on these field data, we calibrate the model and show the resulting attitude dynamics. We study the sensibility of the model to the co-responsibility factor.

International audience ; This paper presents a multi-agent model for simulating attitude formation and change based on perception and communication in the context of stabilization operations. The originality of our model comes from (1) attitude computation that evaluates information as part of a history relative to the individual (2) a notion of co-responsibility forattitude attribution. We present a military scenario of French operations in Afghanistan along with polls results about the opinion of citizen toward present Forces. Based on these eld data, we calibrate the model and show the resulting attitude dynamics. We study the sensibility of the model to the co-responsibility factor.

International audience ; This paper presents a multi-agent model for simulating attitude formation and change based on perception and communication in the context of stabilization operations. The originality of our model comes from (1) attitude computation that evaluates information as part of a history relative to the individual (2) a notion of co-responsibility forattitude attribution. We present a military scenario of French operations in Afghanistan along with polls results about the opinion of citizen toward present Forces. Based on these eld data, we calibrate the model and show the resulting attitude dynamics. We study the sensibility of the model to the co-responsibility factor.

International audience ; In the context of military training for stabilizationoperation of a crisis zone with civilian population, understandingthe formation of attitude and its dynamics is a key issue. Thispaper presents a multi-agent model for simulating attitude forma-tion and change based on individuals perception of informationand its diffusion through communication. We represent theattitude as object-evaluation associations of varying strength pro-posed by Fazio [1]. Individuals observe military operations. Theyexchange and revise beliefs about social objects depending onmultiple criteria deriving from social psychology theories. Theycompute their attitude value based on analytic assessment of thesebeliefs. We illustrate, through several simulation experiments, therole of communication on attitude dynamics.

International audience ; In the context of military training for stabilizationoperation of a crisis zone with civilian population, understandingthe formation of attitude and its dynamics is a key issue. Thispaper presents a multi-agent model for simulating attitude forma-tion and change based on individual's perception of informationand its diffusion through communication. We represent theattitude as object-evaluation associations of varying strength pro-posed by Fazio [1]. Individuals observe military operations. Theyexchange and revise beliefs about social objects depending onmultiple criteria deriving from social psychology theories. Theycompute their attitude value based on analytic assessment of thesebeliefs. We illustrate, through several simulation experiments, therole of communication on attitude dynamics.

International audience ; Health care delivery in military conflict, in peacekeeping missionsor in the aftermath of disaster, implies high stress environmentswith danger exposures, life-threatening events and high levels ofwork demand. Crisis and emergency risk communication remainsa real challenge. The rapid response of emergency medical teamshas an important role to play in preventing serious adverse events.During critical events, medical errors can be related to human or systemfactors, including ineffective team leadership, non-standardizedteam communication, a lack of global situation awareness, pooruse of resources and inappropriate triage and prioritization. TheVICTEAMS project aims at building a virtual environment for trainingrescue team leaders to non-technical skills. Depending on theleader abilities, the virtual characters playing the subordinates (e.g.nurses) reproduce a variety of behaviors like erroneous actions,followership attitudes or stress-based behaviors. A pedagogical directortailors the difficulty according to the dynamic profile of thelearner.

International audience ; Health care delivery in military conflict, in peacekeeping missionsor in the aftermath of disaster, implies high stress environmentswith danger exposures, life-threatening events and high levels ofwork demand. Crisis and emergency risk communication remainsa real challenge. The rapid response of emergency medical teamshas an important role to play in preventing serious adverse events.During critical events, medical errors can be related to human or systemfactors, including ineffective team leadership, non-standardizedteam communication, a lack of global situation awareness, pooruse of resources and inappropriate triage and prioritization. TheVICTEAMS project aims at building a virtual environment for trainingrescue team leaders to non-technical skills. Depending on theleader abilities, the virtual characters playing the subordinates (e.g.nurses) reproduce a variety of behaviors like erroneous actions,followership attitudes or stress-based behaviors. A pedagogical directortailors the difficulty according to the dynamic profile of thelearner.