An Evolving Agent with EVOLP

Logic programming has often been considered less than ad-equate for modelling the dynamics of knowledge changing over time.Evolving Logic Programs (EVOLP) has been recently proposed as a sim-ple though quite powerful extension of logic programming, which allowsfor modelling the dynamics of knowledge bases expressed by programs,and illustrate its usage in mo delling agents whose specifications maydynamically change. From the syntactical point of view, evolving pro-grams are just generalized logic programs (i.e. normal LPs plus defaultnegation in rule heads too), extended with (possibly nested) assertions,whether in heads or bodies of rules. From the semantical point of view,a model-theoretic characterization is offered of th e possible evolutions ofsuch programs. These evolutions arise both from self (i.e. internal to theagent) updating, and from external updating originating in the environ-ment. In this paper we illustrate t he usage and power of EVOLP, andits ability to model agents’ specifications, by elaborating on variationsin the modelling of a Personal Assistant Agent for e-mail management.