TY - GEN
T1 - How people talk when teaching a robot
AU - Kim, Elizabeth S.
AU - Leyzberg, Dan
AU - Tsui, Katherine M.
AU - Scassellati, Brian
N1 - Copyright:
Copyright 2012 Elsevier B.V., All rights reserved.
PY - 2008
Y1 - 2008
N2 - We examine affective vocalizations provided by human teachers to robotic learners. In unscripted one-on-one interactions, participants provided vocal input to a robotic dinosaur as the robot selected toy buildings to knock down. We find that (1) people vary their vocal input depending on the learner's performance history, (2) people do not wait until a robotic learner completes an action before they provide input and (3) people näively and spontaneously use intensely affective vocalizations. Our findings suggest modifications may be needed to traditional machine learning models to better fit observed human tendencies. Our observations of human behavior contradict the popular assumptions made by machine learning algorithms (in particular, reinforcement learning) that the reward function is stationary and pathindependent for social learning interactions. We also propose an interaction taxonomy that describes three phases of a human-teacher's vocalizations: direction, spoken before an action is taken; guidance, spoken as the learner communicates an intended action; and feedback, spoken in response to a completed action.
AB - We examine affective vocalizations provided by human teachers to robotic learners. In unscripted one-on-one interactions, participants provided vocal input to a robotic dinosaur as the robot selected toy buildings to knock down. We find that (1) people vary their vocal input depending on the learner's performance history, (2) people do not wait until a robotic learner completes an action before they provide input and (3) people näively and spontaneously use intensely affective vocalizations. Our findings suggest modifications may be needed to traditional machine learning models to better fit observed human tendencies. Our observations of human behavior contradict the popular assumptions made by machine learning algorithms (in particular, reinforcement learning) that the reward function is stationary and pathindependent for social learning interactions. We also propose an interaction taxonomy that describes three phases of a human-teacher's vocalizations: direction, spoken before an action is taken; guidance, spoken as the learner communicates an intended action; and feedback, spoken in response to a completed action.
KW - Experimentation
KW - Human factors
UR - http://www.scopus.com/inward/record.url?scp=67650686482&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=67650686482&partnerID=8YFLogxK
U2 - 10.1145/1514095.1514102
DO - 10.1145/1514095.1514102
M3 - Conference contribution
AN - SCOPUS:67650686482
SN - 9781605584041
T3 - Proceedings of the 4th ACM/IEEE International Conference on Human-Robot Interaction, HRI'09
SP - 23
EP - 30
BT - Proceedings of the 4th ACM/IEEE International Conference on Human-Robot Interaction, HRI'09
T2 - 4th ACM/IEEE International Conference on Human-Robot Interaction, HRI'09
Y2 - 11 March 2009 through 13 March 2009
ER -