TY - GEN
T1 - Learning effects of robot actions using temporal associations
AU - Cohen, P. R.
AU - Sutton, C.
AU - Burns, B.
N1 - Publisher Copyright:
© 2002 IEEE.
PY - 2002
Y1 - 2002
N2 - Agents need to know the effects of their actions. Strong associations between actions and effects can be found by counting how often they co-occur. We present an algorithm that learns temporal patterns expressed as fluents, i.e. propositions with temporal extent. The fluent-learning algorithm is hierarchical and unsupervised. It works by maintaining co-occurrence statistics on pairs of fluents. In experiments on a mobile robot, the fluent-learning algorithm found temporal associations that correspond to effects of the robot's actions.
AB - Agents need to know the effects of their actions. Strong associations between actions and effects can be found by counting how often they co-occur. We present an algorithm that learns temporal patterns expressed as fluents, i.e. propositions with temporal extent. The fluent-learning algorithm is hierarchical and unsupervised. It works by maintaining co-occurrence statistics on pairs of fluents. In experiments on a mobile robot, the fluent-learning algorithm found temporal associations that correspond to effects of the robot's actions.
UR - https://www.scopus.com/pages/publications/84962033929
UR - https://www.scopus.com/pages/publications/84962033929#tab=citedBy
U2 - 10.1109/DEVLRN.2002.1011807
DO - 10.1109/DEVLRN.2002.1011807
M3 - Conference contribution
AN - SCOPUS:84962033929
T3 - Proceedings - 2nd International Conference on Development and Learning, ICDL 2002
SP - 96
EP - 101
BT - Proceedings - 2nd International Conference on Development and Learning, ICDL 2002
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2nd International Conference on Development and Learning, ICDL 2002
Y2 - 12 June 2002 through 15 June 2002
ER -