CONF Moerland-95.1/IDIAP The Effects of Optical Thresholding in Backpropagation Neural Networks Moerland, Perry Fiesler, Emile Saxena, Indu Fogelman-Soulié, F. Ed. Gallinari, P. Ed. ENNS - Proceedings of the International Conference on Artificial Neural Networks (ICANN'95 and NeuroNimes'95) Paris, France 2 339-343 2-910085-19-8 1995 EC2 & Cie Paris La Défense, France Sigmoid-like activation functions implemented in analog hardware differ in various ways from the standard sigmoidal function as they are asymmetric, truncated, and have a non-standard gain. It is demonstrated how one can adapt the backpropagation learning rule to compensate for these non-standard sigmoids as available in hardware. This method is applied to multilayer neural networks with all-optical forward propagation and liquid crystal light valves (LCLV) as optical thresholding devices. In this paper the results of software simulations of a backpropagation neural network with five different LCLV activation functions are presented and it is shown that the adapted learning rule performs well with these LCLV curves