Abstract Concept Learning in Cognitive Robots

Abstract Purpose of Review Understanding and manipulating abstract concepts is a fundamental characteristic human intelligence that currently missing in artificial agents. Without it, the ability these robots to interact socially with humans while performing their tasks would be hindered. However, what needed empower our such capability? In this article, we discuss some recent attempts on cognitive robot modeling underpinned by neurophysiological principles. Recent Findings For advanced learning concepts, an agent needs (robotic) body, because concrete are considered continuum, can learned linking them embodied perceptions. Pioneering studies provided valuable information about simulation demonstrated value robotics approach study aspects cognition. Summary There few successful examples models knowledge based connectionist probabilistic techniques. concept limited at narrow tasks. To make further progress, argue closer collaboration among multiple disciplines required share expertise co-design future studies. Particularly important create benchmark datasets behavior.