Sliding mode algorithm for training multilayerartificial neural networks
G.G. Parma, Benjamim Rodrigues de Menezes, Antônio P. Braga · Electronics Letters · 1998
On-line learning algorithms for artificial neural networks (ANNs) are expected to adapt network parameters in order to face new control situations. A new on-line learning algorithm, based on sliding mode control (SMC) is presented. The results show that ANN inherits some of the advantages of SMC: high speed of learning and robustness.