Understanding agent systems
Over the last decade, agent-based systems have moved from a niche area of interest in Al into the mainstream. Among others, the reason for the growth of their popularity refers to the progress made in complementary technologies, the most dramatic of which has been probably the emergence of the World Wide Web. Nowadays, we see enormous commercial interest in agent products and agent systems are becoming a mature technology. This book provides an abstract architecture for such agents and shows how their increased reasoning capabilities enable them to make qualified social decisions with respect to multi-agent plans containing actions to be performed by others. It develops a framework for various definitions, architectures, designs, theories and systems to be related within a single coherent whole. A recurrent theme is the lack of aggreement over what it is that actually constitutes an agent. This results from the definition of agents as objects with an ascribed set of goals. Goal adoption represents the key issue underlying multi-agent systems. Once an agent has adopted the goal of another, an essential relationship is created between the two agents, with one acting on behalf of the other. The authors argue that in order to be effective in multi-agent systems; agents need also access to the information about agent relationships. Agents should be further able to recognise these relationships in the system of which they are a part. The model developed in this book allows to reason about how and why goal adoption takes place and provides an understanding of the origin of these relationships -- how and when they are initiated, manipulated, or broken and how any changes may impact on other relationships. The whole text is divided into 9 chapters. Chapter 1 outlines the key concepts of agent-based systems. Chapter 2 provides the SMART agent framework using the Z specification language. The Z language was developed in the late 1970s and is based on set theory and first-order predicate calculus. The relationships between agents are discussed in Chapter 3. Chapter 4 describes the ways in which inter-agent relationships can be created and destroyed. In Chapter 5, the SMART framework is further developed to include different dimensions of agents necessary for their effective operation in multi-agent. To illustrate the application of the SMART framework to a range of diverse systems and theories, Chapters 6--8 are devoted to case studies. The concluding Chapter 9 reviews the overall contribution of the book, and evaluates the applicability of SMART. An introduction to the syntax of the Z language is included as an Appendix. The book is written clearly and contains many illustrative figures, supporting graphs and sufficient references. It provides a sound conceptual framework for describing and organising the landscape of agent-based systems. Its authors do not constrain the terminology just to rigid definitions, but provide an encompassing infrastructure useful to understand the nature of different agent systems. Therefore, this text represents a good source of applicable ideas for researchers, students, and professionals in industry interested in advanced applications of intelligent agents and multi-agent systems.
- scientific article; zbMATH DE number 2020409
- A conceptual framework for agent definition and development
- The agent modeling language -- AML. A comprehensive approach to modeling multi-agent systems
- Communication and cooperation in agent systems. A pragmatic theory
- scientific article; zbMATH DE number 978764
- Design of intelligent multi-agent systems. Human-centredness, architectures, learning and adaptation
- Unifying agent systems
- Communication and cooperation in agent systems. A pragmatic theory
- Inter-agent connections in multi-agent systems
- Architecture-based design of multi-agent systems. Foreword by Len Bass.
- The agent modeling language -- AML. A comprehensive approach to modeling multi-agent systems
- Agent Transportation Layer Adaptation System
- Towards pervasive intelligence: reflections on the evolution of the Agent Factory framework
- scientific article; zbMATH DE number 1333476 (Why is no real title available?)
- scientific article; zbMATH DE number 2020409 (Why is no real title available?)
- On formal modeling of agent computations
- Formal specification of agent behaviour through environment scenarios
- Modelling agents as observable sources
- Programming Multi-Agent Systems in AgentSpeak usingJason
- Modeling and shadowing paraconsistent BDI agents
- Agent technology from a formal perspective.
This page was built for publication: Understanding agent systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2757864)
