Introduction to the special issue on autonomous agents for agent-based modeling

Agent-basedmodeling (ABM) is a general approach to understanding social systems through simulation of interacting agents. An agent-based model is by definition a model of a multi-agent system (MAS), yet historically the ABM and MAS research communities have proceeded on nearly independent tracks. This special issue bridges the MAS and ABM communities, by collecting research contributions that serve goals of both fields. Combining the benefits of ABM and MAS perspectives poses many technical and conceptual challenges, as are illustrated in the papers composing this issue. Validation of ABM models is one such challenge. Zhang et al. [6] address this by proposing a data-driven ABM framework inwhich an individual behaviormodel is acquired bymachine learning techniques, deployed in multi-agent systems and validated using a sequence of collective adoption decisions. They apply this framework to forecasting individual and aggregate residential rooftop solar adoption in San Diego county and demonstrate that the resulting agent-based model successfully forecasts solar adoption trends and provides a meaningful quantification of uncertainty about its predictions. Singh et al. [4] address a key integration challenge of coupling event-based BDI systems to time-stepped ABM systems by presenting a framework that allows belief-desire-intention (BDI) cognitive agents to be embedded in an ABM system. Architecturally, this means that BDI is used to model the brains of agents whose bodies exist in an ABM system. The architecture is flexible in that the ABM can still have non-BDI agents in the simulation, and theBDI-side can have agents that do not have a physical counterpart (such as an organization).