Abstract
Joint attention related behaviors (JARBs) are some of the most important and basic cognitive functions for establishing successful communication in human interaction. It is learned gradually during the infant's developmental process, and enables the infant to purposefully improve his/her interaction with the others. To adopt such a developmental process for building an adaptive and social robot, previous studies proposed several contingency evaluation methods, by which an infant robot becomes able to sequentially learn some primary social skills. These skills included gaze following and social referencing, and could be acquired through interacting with a human caregiver model in a computer simulation. However, to implement such methods to a real-world robot, two major problems, that were not addressed in the previous research, have remained unresearched: (1) dependency of histogram of the observed events by the robot to each other, which increases the error of the internal calculation and consequently decreases the accuracy of contingency evaluation; and (2) unsynchronized teaching/learning phase of the teaching-caregiver and the learning-robot, which leads the robot and the caregiver not to understand the suitable timing for the learning and the teaching, respectively. In this paper, we address these two problems, and propose two algorithms in order to solve them: (1) exclusive evaluation of policies (XEP) for the former, and (2) ostensive-cue sensitive learning (OsL) for the latter. To show the effect of the proposed algorithms, we conducted a real-world human-robot interaction experiment with 48 subjects, and compared the performance of the learning robot with/without proposed algorithms. Our results show that adopting proposed algorithms improves the robot's performance in terms of learning efficiency, complexity of the learned behaviors, predictability of the robot, and even the result of the subjective evaluation of the participants about the intelligence of the robot as well as the quality of the interaction.
Highlights
Joint attention related behaviors (JARBs) include basic social skills, such as following the gaze of others, pointing, intention sharing, and social referencing
We proposed two novel algorithms to improve the performance of the social skill learning of an infant robot during interaction with a human caregiver: namely the Ostensive-cue sensitive Learning (OsL) and the Exclusive Evaluation of Policies (XEP) algorithms
The ostensive-cue sensitive learning (OsL) was inspired by the natural pedagogy of the human being and proposed a synchronized weighted learning mechanism based on the ostensive signals of the caregiver
Summary
Joint attention related behaviors (JARBs) include basic social skills, such as following the gaze of others, pointing, intention sharing, and social referencing. Humans gradually learn these social skills during their developmental process in infancy and childhood (Scaife and Bruner, 1975; Adamson, 1995; Corkum and Moore, 1995), and become able to establish interaction with others. Owing to the important role of such behaviors in achieving successful communication with humans, some robotic research has focused on the study of JARBs in the development of communicative robots (Imai et al, 2003; Breazeal, 2004; Kanda et al, 2004; Kaplan and Hafner, 2006). Some of these research has been done on proposing learning mechanisms based on the intrinsic motivation of the robot that enables open-ended development (Oudeyer et al, 2007; Barto, 2013; Nehmzow et al, 2013), and some on dynamic Bayesian networks to evaluate the contingency of the observed events, which enables the robot to plan suitable action(s) to achieve its goal utilizing the evaluated contingency (Degris et al, 2006; Jonsson and Barto, 2007; Mugan and Kuipers, 2012)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
