Autonomous decision with curiosity and dynamic goal setting

In this paper, a model of behavior emergence with curiosities is proposed. The proposed model is applied to an agent in “Optimal Patch Use Problem”. The curious agent dynamically changes its action strategy, according to its degree of satiation and its environment at a given point in time. The agent efficiently searches for food by virtue of its diversive and specific curiosities. The effectiveness of the proposed model is verified in simulations.