Multi agent systems are an important research area in computer science. They can be viewed as a computational paradigm (like object-oriented and component-based paradigms) to specify, design, and implement software systems.
Multi agent systems aim at developing systems to solve complex tasks in terms of interacting computational entities, called agents. The agents are usually assumed to be heterogeneous, autonomous, and self-interested, while they can communicate and co-ordinate their activities to achieve their objectives.
Multi agent systems can be used to develop applications that are inherently distributed, large, open, and heterogeneous. Examples of such applications are electronic commerce systems such as online auctions, electronic institutions, market places, and organizations.
Multi agent systems are based on many ideas from various scientific fields such as philosophy, psychology, economics, physics, and organizational and management sciences. Some of multi agent issues are: how agents represent and reason about their (mental) states (beliefs, desires, intentions - BDI) and their goals and plans, how they communicate with each other, how they co-ordinate their activities, and what are characteristics of multi agent applications.
We need new systems that can become our partners in expanding the horizon of human cognition to help us navigate the increasing complexity of our globally interconnected world.
Until now, most electronic computers have been based on the “calculating” paradigm.
ATIA's expanding technology frontiers are providing us with:
The opportunity to build a new class of systems that can learn from both structured and unstructured data
Finding important correlations
Ccreate hypotheses for these correlations
Suggest and measure actions to enable better outcomes for users.
Systems with these capabilities will transform our view of computers from “calculators” to “machines that learn”, a shift that will radically alter our expectations of what computing is and the nature of problems it should help us solve. These systems will impact virtually every sector of the economy, enabling applications and services that will range from preventing fraud and providing better security, to helping launch new products or to improving medical diagnosis.
Achieving this level of performance will require advances (and sometimes breakthroughs) in learning algorithms and architectures, expanded data input and output modalities (e.g. the ability to process text, graphs, images, video, sound, and other sensory information).
ATIA’s research is focused on these multi agent systems using them for a subtle diagnostic system of complex multi disciplinary domains.
Consider a medical psychological problem about which physical and/or psychological patient data are available. The data span several scientific (sub)disciplines. How can you combine these in such a way that the result is a workable diagnosis of the state of the patient? Of course a medicine (or even several medical specialists) could deliberate with a psychologist about a patient. But can that be done more effectively?
ATIA is working on an intelligent system that facilitates consulting and tuning. The purpose is to develop a (heterogeneous) multi agent system (HeMAS) in which the separate agents each have their own reasoning method. The system is called heterogeneous not only because of the diversity of reasoning methods, but also because the knowledge that the agents reason with comes from a diversity of domains.
The separate agents have to ‘deliberate’ with each other in order to reach a well considered final judgment; a discussion panel of artificial intelligent experts!
Perhaps it sounds simple, but of course it isn't. There are a lot of questions to be answered
How do agents exchange their knowledge?
How do they combine their knowledge?
How do they try to convince each other?
This is an enormous challenge for the today’s research in artificial intelligence and agent technology in particular.