TY - GEN
T1 - Feedback stabilization using two-hidden-layer nets
AU - Sontag, Eduardo D.
PY - 1991
Y1 - 1991
N2 - The global stabilization of nonlinear systems is considered by means of state feedback laws which can be implemented using feedforward neural networks. The objective is not to provide a practical stabilization technique, but rather to explore the capabilities and the ultimate limitations of alternative network architectures. Contrary to what might have been expected from the well-known representation theorems, three-layer (also called single hidden layer) nets are not sufficient for stabilization, but four-layer nets are enough, assuming that threshold processors are used.
AB - The global stabilization of nonlinear systems is considered by means of state feedback laws which can be implemented using feedforward neural networks. The objective is not to provide a practical stabilization technique, but rather to explore the capabilities and the ultimate limitations of alternative network architectures. Contrary to what might have been expected from the well-known representation theorems, three-layer (also called single hidden layer) nets are not sufficient for stabilization, but four-layer nets are enough, assuming that threshold processors are used.
UR - http://www.scopus.com/inward/record.url?scp=0026382199&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0026382199&partnerID=8YFLogxK
U2 - 10.23919/acc.1991.4791486
DO - 10.23919/acc.1991.4791486
M3 - Conference contribution
AN - SCOPUS:0026382199
SN - 0879425652
SN - 9780879425654
T3 - Proceedings of the American Control Conference
SP - 815
EP - 820
BT - Proceedings of the American Control Conference
PB - Publ by American Automatic Control Council
T2 - Proceedings of the 1991 American Control Conference
Y2 - 26 June 1991 through 28 June 1991
ER -