Determining the Envelope of Emergent Agent Behaviour via Architectural Transformation

Abstract

Abstract. In this paper we propose a methodology to help analyse tendencies in MAS to complement those of simple inspection, Monte Carlo and syntactic proof. We suggest an architecture that allows an exhaustive model-based search of possible system trajectories in significant fragments of a MAS using forward inference. The idea is to identify tendencies, especially emergent tendencies, by automating the search through possible parameterisations of the model and the choices made by the agents. Subsequently, a proof of these tendencies could be attempted over all possible conditions using syntactic proof procedures. Additionally, we propose a computational procedure to help implement this. The strategy consists of: unencapsulating the MAS so as to reveal the maximum information about logical dependencies in the system. This information is maximised by splitting the transition rules by time intervals and some parameters. An example applying this procedure is exhibited which 'compiles' the rules into this form. In the example the exploration of possibilities is speeded up by a factor of 14. This makes possible the complete exploration of model behaviour over a range of parameterisations and agent choices.

Defining and addressing the risk of undesirable emergent properties

“…Methods to predict emergent properties are mainly based on pattern matching [23][24][25][26] and simulation [6,[27][28][29][30][31]. Honour [11] identifies and discusses methods that can be used to evaluate emergent properties.…”