In recent years we have seen substantial progress being made in the area of interactive adaptive systems. Nevertheless, there are still a number of challenges that need to be addressed before adaptivity becomes a widely employed "instrument" in the design and implementation of adaptive systems. Firstly, there is still little empirically validated evidence as to what adaptation methods and techniques work, for what categories of users, and in what contexts of use. Secondly, although formulating design goals for adaptivity is relatively easy, evaluating against such goals has proven to be a complicated and elusive target.
This presentation will introduce self-regulation and proceed to discuss how its employment may facilitate addressing these two problems. Self-regulation refers to an adaptive system's capacity to perform self-evaluation against design objectives, formulate and accumulate new adaptation knowledge, and use that knowledge to modify its own adaptive behaviour. The presentation will provide an introduction to the concept of self-regulation, and the requirements it places on the adaptation infrastructure, as well as on the design and implementation process itself. It will then discuss the opportunities arising from its employment as a support tool for the early design phases of adaptation, as well as how it can help attain a degree of self-evolution in deployed adaptive systems.