Collaborative Agent Tuning

Ambient intelligence envisages a world saturated with sensors and other embedded computing technologies, operating transparently, and accessible to all in a seamless and intuitive manner. Intelligent agents of varying capabilities may well form the essential constituent entities around which this vision is realized, potentially resulting in multiple, large-scale, open MultiAgent Systems. However, the practical realization of this vision will severely exacerbate the complexity of existing software solutions and supporting network infrastructures, a problem that autonomic computing was originally conceived to address. Thus we can conjecture that the incorporation of autonomic principles into the design of Multi-Agent Systems is indeed a desirable objective. As an illustration of how this may be achieved, a strategy termed Collaborative Agent Tuning is described, which seeks to optimise agent performance on computationally limited devices. A classic mobile computing application is used to illustrate the principles involved.