Artificially intelligent agents increasingly collaborate with humans in human-agent teams. Timely proactive sharing of relevant information within the team contributes to the overall team performance. This paper presents a machine learning approach to proactive communication in AI-agents using contextual factors. Proactive communication was learned in two consecutive experimental steps: (a) multi-agent team simulations to learn effective communicative behaviors, and (b) human-agent team experiments to refine communication suitable for a human team member. Results consist of proactive communication policies for communicating both beliefs and goals within human-agent teams. Agents learned to use minimal communication to improve team performance in simulation, while they learned more specific socially desirable behaviors in the human-agent team experiment
DOCUMENT
In this paper we explore the influence of the physical and social environment (the design space) son the formation of shared understanding in multidisciplinary design teams. We concentrate on the creative design meeting as a microenvironment for studying processes of design communication. Our applied research context entails the design of mixed physical–digital interactive systems supporting design meetings. Informed by theories of embodiment that have recently gained interest in cognitive science, we focus on the role of interactive “traces,” representational artifacts both created and used by participants as scaffolds for creating shared understanding. Our research through design approach resulted in two prototypes that form two concrete proposals of how the environment may scaffold shared understanding in design meetings. In several user studies we observed users working with our systems in natural contexts. Our analysis reveals how an ensemble of ongoing social as well as physical interactions, scaffolded by the interactive environment, grounds the formation of shared understanding in teams. We discuss implications for designing collaborative tools and for design communication theory in general.
MULTIFILE
Accurate modeling of end-users’ decision-making behavior is crucial for validating demand response (DR) policies. However, existing models usually represent the decision-making behavior as an optimization problem, neglecting the impact of human psychology on decisions. In this paper, we propose a Belief-Desire-Intention (BDI) agent model to model end-users’ decision-making under DR. This model has the ability to perceive environmental information, generate different power scheduling plans, and make decisions that align with its own interests. The key modeling capabilities of the proposed model have been validated in a household end-user with flexible loads
DOCUMENT
