A Formal Model of Agent Interaction Based on MASQ

Awareness is a concept that has been frequently studied in the context of Computer Supported Cooperative Work. However, other fields of computer science can benefit from this concept. Recent research in the multi-agent systems field has highlighted the relevance of complex interaction models such as multi-party communication and context awareness for simulation and adaptive systems. In this article, we present a generic interaction model that enables to use these different models in a standardized way. Emerging as a first-order abstraction, the environment, in the sense of a common medium for the agents, is a suitable paradigm to support the agents' awareness. We present an operational model, called Environment as Active Support of Interaction, to take into account all the agents that can be interested in a communication. This model is then extended for the regulation of multiagent systems interactions. Priority policies are given to manage the rules governing the context (un-)awareness of the agents. We also present a new AUML connector to create protocols that take into account the agent awareness to implement proactive behaviour, and several communication scenarios are proposed to show practical applications of this model.

Digitalization and the use of autonomous systems challenge companies in new ways, particularly concerning the automatic distribution of tasks. Various approaches are being discussed in research, specifically in the field of multi-agent systems. However, there does not seem to be a universally applicable and generally accepted solution. Due to the multitude of different approaches and the associated challenges, traceability is of particular interest and requires further analysis. This paper reviews the traceability of different negotiation-based approaches for task allocation in multi-agent systems. We conducted a structured literature review and implemented a prototype using an existing workflow engine for the approach we identified as most suitable, applying a weighted scoring model. The evaluation of the implemented demonstrator, performed both with and without restarting negotiations, indicates that a traceable distribution of tasks with a negotiation mechanism is feasible.

The field of multi-agent dynamic systems is an inter-disciplinary research field that has become very popular in the recent years in parallel with the significant interest in the practical applications of such systems in various areas including robotics. In this article we give a relatively short review of this field from the system dynamics and control perspective. We first focus on mathematical modelling of multi-agent systems paying particular attention on the agent dynamics models available in the literature. Then we present a number of problems on coordination and control of multi-agent systems which have gained significant attention recently and various approaches to these problems. Relevant to these problems and approaches, we summarize some of the recent results on stability, robustness, and performance of multi-agent dynamic systems which appeared in the literature. The article is concluded with some remarks on the implementation and application side of the control designs developed for multi-agent systems.

The main challenge that artificial intelligence research is facing nowadays is how to guarantee the development of responsible technology. And, in particular, how to guarantee that autonomy is responsible. The social fears on the actions taken by AI can only be appeased by providing ethical certification and transparency of systems. (See for instance the Barcelona declaration https://www.iiia.csic.es/barcelonadeclaration/) However, this is certainly not an easy task. As we very well know in the multiagent systems field, the prediction accuracy of system outcomes has limits as multiagent systems are actually examples of complex systems. And AI will be social, there will be thousands of AI systems interacting among themselves and with a multitude of humans; AI will necessarily be multiagent. Although we cannot provide complete guarantees on outcomes, we must be able to define with accuracy what autonomous behaviour is acceptable (ethical), to provide repair methods for anomalous behaviour and to explain the rationale of AI decisions. Ideally, we should be able to guarantee responsible behaviour of individual AI systems by construction. I understand by an ethical AI system one that is capable of deciding what are the most convenient norms, abide by them and make them evolve and adapt. The area of multiagent systems has developed a number of theoretical and practical tools that properly combined can provide a path to develop such systems, that is, provide means to build ethical-by-construction systems: agreement technologies to decide on acceptable ethical behaviour, normative frameworks to represent and reason on ethics, and electronic institutions to operationalise ethical interactions. Along my career, I have contributed with tools on these three areas. In this keynote, I will describe a methodology to support their combination that incorporates some new ideas from law, and organisational theory.

Choosing social laws for multi-agent systems: Minimality and simplicity

In this paper, the paradigm of rationality is examined from the point of view of designing, and possibly implementing, agents endowed with the capacity to interact in a multi-agent environment. The work that will be presented originates from a cross-fertilization between the fields of social simulation and Multi-Agent Systems (MAS) research. In the MAS field, models of intelligent agents in a multi-agent context have been formalized and implemented. The question then arises as to whether the model of rational agents may contribute fruitfully to construct and implement multi-agents systems.

A New Approach In Learning For Intelligent Multi Agent Systems

In Multi-Agent System (MAS), developers concentrate on creating design models and evolving them, from higher level models to lower level models, in several steps. Considerable part of MAS implementations is automatically produced from the design models. If a design model contains faults, they are passed to the generated implementations. Practical model validation techniques are required to discover and delete faults in abstract design models. In this paper, we introduce a formal approach for MAS design testing. It specifies a testing process that complements Multi-agent Systems Engineering (MaSE) methodology and strengthens the mutual relationship between UML and MAS. Besides, it defines a structured and comprehensive testing process for engineering software agents at the design level by providing a systematic way of converting the MAS design models to UML design diagram. Then a Petri Net (PN) diagram is generated from the UML models to simulate the behavior of the MAS system. Finally, because Petri Nets (PNs) are formal models, their analysis techniques can be applied to automatic MAS behavioral testing.

Social laws have proved to be a powerful and theoretically elegant framework for coordination in multi-agent systems. Most existing models of social laws assume that a designer is attempting to produce a set of constraints on agent behaviour which will ensure that some single overall desirable objective is achieved. However, this represents a gross simplification of the typical situation, where a designer may have multiple (possibly conflicting) objectives, with different priorities. Moreover, social laws, as well as bringing benefits, also have implementation costs: imposing a social law often cannot be done at zero cost. We present a model of social laws that reflects this reality: it takes into account both the fact that the designer of a social law may have multiple differently valued objectives, and that the implementation of a social law is not costneutral. In this setting, designing a social law becomes an optimisation problem, in which a designer must take into account both the benefits and costs of a social law. We investigate the issue of representing a designer's objectives, characterise the complexity of the optimal social law design problem, and consider possible constraints that lead to reductions in computational complexity. We then show how the problem of designing an optimal social law can be formulated as an integer linear program.

In the field of multi-agent systems, a lot of attention has been focused lately on investigating various architectures and methodologies that promote effective organization and coordination within large-scale, complex, distributed systems. Specifically, the interest is in developing approaches that can be implemented within multi-agent systems to produce desirable emergent behavior that coordinates individual actors in a system competing for resources such as bandwidth, computing power, and data. This extended abstract outlines polymorphic self-* agents that evolve a core set of roles and behavior based on environmental cues. The agents adapt these roles based on the changing demands of the environment. The design combines strategies from game theory, stigmergy, and other biologically inspired models.

A major challenge in the field of multi-agent systems is to enable autonomous agents to allocate tasks efficiently. In the context of massive multi-agents system (MMAS) which is characterised by a large number of dynamic and heterogeneous agents, traditional tasks allocations approaches based on the negotiation between agents, or a single allocator agent, proved impracticality. In this paper, we propose a decentralised and scalable approach for complex task allocation pro cooperative MMAS. Our approach is based on the idea of grouping agents according to their capacities to execute sub-tasks. Thus, the approach is based on two steps: 1 hierarchical organisation of agent groups using Galois sub-hierarchy, imminent from the formal concepts analysis approach (FCA) 2 computing the optimal allocation. These two steps are fully distributed among agents with the minimum of communication and a grant of finding the optimal allocation in a polynomial time. Further, this paper extends our last approach by distributing the global allocation process among all agents. It provides a solution based on cooperation among agents. This solution prohibits generation of conflicts. It is based on the idea that each agent has to pick out its own sub-task.

A major challenge in the field of Multi-Agent Systems is to enable autonomous agents to allocate tasks efficiently. In previous work, we have developed a decentralized and scalable method for complex tasks allocation for Massive Multi-Agent System (MMAS) based on two steps: 1) hierarchical organization of agent groups using Formal Concepts Analysis approach (FCA), 2) computing the optimal allocation. The first step is computed by one agent named global allocator that computes Galois lattice representing the hierarchical structure of agent groups. Then, it simplifies the completed lattices by pruning unnecessary groups. The second step distributes the tasks allocation process among all agent groups. Nevertheless, the hierarchical organization process is still centralized. Moreover, generation of Galois lattice composed by all concepts (2 min (|O |,|A |) concepts in the worst cases. Where 0 and A means, respectively, the set of objects and the set of attributs) and then simplification of the hierarchy of such size are not useful. This paper extends our last approach to distribute the organization process of agent groups among all agents by providing extension to the Pulton algorithm that generates Galois Sub-Hierarchy which is a polynomial size representation of a concept lattice. This decentralized self-organization of agents provides a flexible infrastructure for agents' dynamicity in MMAS.

Nowadays, in order to face new challenges, extended enterprises must be up to date to the new strategic, economic and organizational structures. Consequently, intelligent software based on agent technology emerges to improve system design, and to increase enterprise competitive position as well. The competitiveness is based on the cooperation. Thus, within this cooperation, conflicts may arise. The automated negotiation plays a key role to look for a common agreement. Argumentation theory has become an important topic in the field of Multi-Agent Systems and especially in the negotiation problem. In this paper, first, an overview of an already proposed model MAIS-E2 (Multi-Agent Information System for an Extended Enterprise) is presented. Then an argumentation based negotiation framework: Relationship-Role and Interest Based Negotiation (R2-IBN) framework is presented, and within this framework, we focused mainly on, argument generation module via inference rules and argument selection module via fuzzy logic.

The paper considers the approach of using multi-agent models in simulation and decision making of complex systems. Furthermore, an overview of research in the field of multi-agent systems and software platforms for their implementation is carried out. In addition, an analysis of the applicability of intelligent multi-agent systems in the tasks of modelling, optimization and management of energy facilities is performed. This study shows that power industry can use intelligent multi-agent systems not only for the tasks of modelling the interaction of subjects as suppliers and consumers of electricity in the wholesale electricity market and other problems of game theory considering a small number of agents’ types. Moreover, intelligent multi-agent systems can be used to model power systems when it is necessary to take into account the deep integration of the systems. This is understood in the work as dependencies at the level of physical processes of generation, transmission, distribution and consumption of electricity, and at the level of interaction between large technological objects and economic entities. At the second level, it is necessary to consider infrastructure linkages and legal regulations for the economic interaction. As an example, the paper considers a hydropower plant. The entity that manages it must take into account not only the features of the station as a complex electrical facility itself, but also the features of the water usage regime, electricity consumption schedules, meteorological factors, etc. In turn, the use of water resources affects the interests of a large number of subjects, which also needs to be reckoned.

In this paper, a simplified reinforcement learning (RL) of identifier-actor-critic architecture is proposed to solve the formation problem of multiagent state-delay system. The dynamics of multi-agent systems includes the uncertainties and state delay, which is more practical in applications. In the multiagent system formation control, the system uncertainties are counteracted through the fuzzy logic system; the state delay is offset by applying a Lyapunov-Krasovskii functional. The updating laws of RL are derived from a simple equation, which is equivalent to the gradient of the HJB equation. Finally, a simulation example is given to demonstrate the satisfactory performance of the proposed method.