TY - GEN
T1 - A modified spreading algorithm for autoassociation in weightless neural networks
AU - Browne, Christopher
AU - de L. Pereira Castro Jr., Joel
AU - Storkey, Amos
PY - 1996
Y1 - 1996
N2 - This paper describes a problem with the conventional Hamming metric spreading algorithm often employed in weightless neural networks. The algorithm can cause incorrect classifications in some cases where a section of the neuron input vector is noise. The conditions under which such error occurs are described and a modified spreading algorithm proposed to overcome this problem has its validity demonstrated theoretically and tested in two practical applications.
AB - This paper describes a problem with the conventional Hamming metric spreading algorithm often employed in weightless neural networks. The algorithm can cause incorrect classifications in some cases where a section of the neuron input vector is noise. The conditions under which such error occurs are described and a modified spreading algorithm proposed to overcome this problem has its validity demonstrated theoretically and tested in two practical applications.
U2 - 10.1007/3-540-61510-5_97
DO - 10.1007/3-540-61510-5_97
M3 - Conference contribution
SN - 978-3-540-61510-1
T3 - Lecture Notes in Computer Science
SP - 569
EP - 574
BT - Artificial Neural Networks — ICANN 96
A2 - von der Malsburg, Christoph
A2 - von Seelen, Werner
A2 - Vorbrüggen, Jan C.
A2 - Sendhoff, Bernhard
PB - Springer Berlin Heidelberg
ER -