TY - GEN
T1 - Learning features and predictive transformation encoding based on a horizontal product model
AU - Zhong, Junpei
AU - Weber, Cornelius
AU - Wermter, Stefan
PY - 2012
Y1 - 2012
N2 - The visual system processes the features and movement of an object in separate pathways, called the ventral and dorsal streams. To integrate this principle in a functional model, a recurrent predictive network with a horizontal product is introduced. Learned in an unsupervised manner, two sets of hidden units represent cells in the ventral and dorsal pathways, respectively. Experiments show that the activity in the ventral-like units persists, given that the same feature appears in the receptive field, whilst the activity in the dorsal-like units shows a fluctuating pattern with different directions of object movements. Moreover, we show that the position information predicts the input's future position taking into account its moving direction due to the direction-selective responses of the dorsal-like units.
AB - The visual system processes the features and movement of an object in separate pathways, called the ventral and dorsal streams. To integrate this principle in a functional model, a recurrent predictive network with a horizontal product is introduced. Learned in an unsupervised manner, two sets of hidden units represent cells in the ventral and dorsal pathways, respectively. Experiments show that the activity in the ventral-like units persists, given that the same feature appears in the receptive field, whilst the activity in the dorsal-like units shows a fluctuating pattern with different directions of object movements. Moreover, we show that the position information predicts the input's future position taking into account its moving direction due to the direction-selective responses of the dorsal-like units.
UR - http://www.scopus.com/inward/record.url?scp=84867685631&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84867685631&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-33269-2_68
DO - 10.1007/978-3-642-33269-2_68
M3 - Conference contribution
AN - SCOPUS:84867685631
SN - 9783642332685
VL - 7552 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 539
EP - 546
BT - Artificial Neural Networks and Machine Learning, ICANN 2012 - 22nd International Conference on Artificial Neural Networks, Proceedings
T2 - 22nd International Conference on Artificial Neural Networks, ICANN 2012
Y2 - 11 September 2012 through 14 September 2012
ER -