In this study, we examine human social interactions within virtual worlds and address the question of how group interactions are affected by the game environment. To investigate this problem, we introduced a set of conversational agents into the social environment of Second Life, a massively multi-player online environment that allows users to construct and inhabit their own 3D world. Our agents were created to be sufficiently lifelike to casual observers, so as not to perturb neighboring social interactions. Using our partitioning algorithm, we separated continuous public chat logs from each region into separate conversations which were used to construct a social network of the participants. Unlike many groups formed in communities and workplaces, groups in Second Life can be rapidly-forming (arising from few interactions), persistent (remaining stable over a long period), and are less affected by socio-cultural influences. In this paper, we analyze regional differences in Second Life by measuring characteristics of the network as a whole, determined from the statistics mined from public conversations in the virtual world, rather than focusing on egocentric actors and their attributes.

We discuss an interaction-based approach to study the coevolution between socio-technical networks, individual behaviors, and contagion processes on these networks. We use epidemics in human population as an example of this phenomenon. The methods consist of developing synthetic yet realistic national-scale networks using a first principles approach. Unlike simple random graph techniques, these methods combine real world data sources with behavioral and social theories to synthesize detailed social contact (proximity) networks. Individual-based models of within-host disease progression and inter-host transmission are then used to model the contagion process. Finally, models of individual behaviors are composed with disease progression models to develop a realistic representation of the complex system in which individual behaviors and the social network adapt to the contagion. These methods are embodied within Simdemics – a general purpose modeling environment to support pandemic planning and response. Simdemics is designed specifically to be scalable to networks with 300 million agents – the underlying algorithms and methods in Simdemics are all high-performance computing oriented methods. New advances in network science, machine learning, high performance computing, data mining and behavioral modeling were necessary to develop Simdemics. Simdemics is combined with two other environments, Simfrastructure and Didactic, to form an integrated cyberenvironment. The integrated cyber-environment provides the end-user flexible and seamless Internet based access to Simdemics. Service-oriented architectures play a critical role in delivering the desired services to the end user. Simdemics, in conjunction with the integrated cyber-environment, has been used in over a dozen user defined case studies. These case studies were done to support specific policy questions that arose in the context of planning the response to pandemics (e.g., H1N1, H5N1) and human initiated bio-terrorism events. These studies played a crucial role in the continual development and improvement of the cyber-environment.

Over the past decade, we have been engaged in an extensive research effort to build virtual humans and applications that use them.  Building a virtual human might be considered the quintessential AI problem, because it brings together many of the key features, such as autonomy, natural communication, sophisticated reasoning and behavior, that distinguish AI systems.  This paper describes major virtual human systems we have built and important lessons we have learned along the way.

Series, held Thursday through Saturday, November 5-7, at he Association for the Advancement of Artificial Intelligence the Westin Arlington Gateway in Arlington, Virginia. The titles of the seven symposia were as follows: (1) Biologically Inspired Cognitive Biologically Inspired Cognitive Architectures Architectures, (2) Cognitive and Metacognitive Cognitive and Metacognitive Educational Systems Educational Systems, (3) Complex Adaptive Complex Adaptive Systems and the Threshold Effect: Views from the Natural Systems and the Threshold Effect: Views and Social Sciences from the Natural and Social Sciences, (4) Manifold Manifold Learning and Its Applications Learning and Its Applications, (5) Multirepresentational Architectures for Human-Level Multirepresentational Architectures for Human-Level Intelligence Intelligence, (6) The Uses of Computational The Uses of Computational Argumentation Argumentation, and (7) Virtual Healthcare Virtual Healthcare Interaction Interaction. An informal reception was held on Thursday, November 5. A general plenary session, in which the highlights of each symposium were presented, was held on Friday, November 6. The challenge of creating a real-life computational equivalent of the human mind requires that we better understand at a computational level how natural intelligent systems develop their cognitive and learning functions. They will behave, variety of disjoined communities and schools of learn, communicate, and "think" as conscious thought that used to speak different languages and beings in general, in addition to being able to perform ignore each other.

We present a knowledge representation framework with an associated run-time support infrastructure that is able to compute, for the benefit of the members of a norm-governed multi-agent system, physically possible and/or permitted actions current at each time, as well as sanctions that should be applied to violations of prohibitions. Experimental results on a benchmark scenario indicate how by distributing norms we can provide run-time support to large-scale, norm-governed multi-agent systems.

Levesque introduced the notion of only-knowing to precisely capture the beliefs of a knowledge base. He also showed how only-knowing can be used to formalize non-monotonic behavior within a monotonic logic. Despite its appeal, all attempts to extend only-knowing to the many agent case have undesirable properties. A belief model by Halpern and Lakemeyer, for instance, appeals to proof-theoretic constructs in the semantics and needs to axiomatize validity as part of the logic. It is also not clear how to generalize their ideas to a first-order case. In this paper, we propose a new account of multi-agent only-knowing which, for the first time, has a natural possible-world semantics for a quantified language with equality. We then provide, for the propositional fragment, a sound and complete axiomatization that faithfully lifts Levesque's proof theory to the many agent case. We also discuss comparisons to the earlier approach by Halpern and Lakemeyer. 

We often deal with dynamic worlds where actions are executed by agents and events may happen. Example of such worlds range from virtual worlds such as the world of a database to robots and humans in physical worlds. To understand the dynamics of such worlds as well as to be able to assert some control over such worlds one needs to reason about the actions and events and how they may change the world. In this invited talk we will present some of the important results in this field and present some future directions. In particular, we will discuss how theories and results from reasoning about actions and change can be combined with theories and results in dynamic epistemic logics to obtain a unified theory of multi-agent actions.

We investigate a class of first-order temporal epistemic logics for the specification of multi-agent systems. We consider well-known properties of multi-agent systems including perfect recall, synchronicity, no learning, unique initial state, and define natural correspondences of these into quantified interpreted systems, the semantics we use to reason about multiagent systems in a first-order setting. Our findings identify several monodic fragments of first-order temporal epistemic logic that we prove to be both sound and complete with respect to the corresponding classes of quantified interpreted systems. The results show that interaction axioms for propositional temporal epistemic logic can be lifted to the monodic fragment.

The applicability conditions of legal Norms regulating computer systems can be modelled in different rules very often refer to these institutional concepts, rather ways, see, for example, (Boella, van der Torre, and than to so called brute facts. To simplify the notation we refer Verhagen 2008). If norms are represented by hard constraints, to the former as constitutive rules, and the latter simply then computer systems are designed to avoid violations.

This is a case-study in knowledge representation. We analyze the 'one hundred prisoners and a lightbulb' puzzle. In this puzzle it is relevant what the agents (prisoners) know, how their knowledge changes due to observations, and how they affect the state of the world by changing facts, i.e., by their actions. These actions depend on the history of previous actions and observations. Part of its interest is that all actions are local, i.e. not publicly observable, and part of the problem is therefore how to disseminate local results to other agents, and make them global. The various solutions to the puzzle are presented as protocols (iterated functions from agent's local states, and histories of actions, to actions). The computational aspect is about average runtime termination under conditions of random ('fair') scheduling. The paper consists of three parts. First, we present different versions of the puzzle, and their solution. This includes a probabilistic version, and a version assuming synchronicity (the interval between prisoners' interrogations is known). The latter is very informative for the prisoners, and allows different protocols (with faster expected termination). Then, we model the puzzle in an epistemic logic incorporating dynamic operators for the effects of information changing events. Such events include both informative actions, where agents become more informed about the non-changing state of the world, and factual changes, wherein the world and the facts describing it change themselves as well. Finally, we give the expected termination results of several protocols when assuming random scheduling. This paper integrates the literature and presents novel contributions. Novel are: Firstly, Protocol 2 and Protocol 4. Secondly, the modelling in dynamic epistemic logic in its entirety - we do not know of a case study that combines factual and informational dynamics in a setting of non-public events, or of a similar proposal to handle asynchronous behaviour in a dynamic epistemic logic. Thirdly, our computational results on Protocol 2 and results from the manuscript from author Wu.