Adaptive Learning Systems
A changing environment can be described by a set of rules, which describes every possible change. In huge and complex spaces this rulesets can become giant. Expert system specially needs rule definitions by human experts. This situation alone prevents the usage of this kind of systems in changing worlds. An additional adaption step can maintain applicability. This adaption step has to incorporate changes to the knowledge base and inference system or find a new solution to the task. The first approach tries to update rules and facts with gained knowledge. One practical application is the machine
learning. The second approach tries to find a solution by searching. As mentioned before in Evolutionary Algorithms the evolutionary algorithms tries to find a solution in huge unknown spaces. To gain convergency the adaption step will have a controlling part.