CONF Moerland-96.3/IDIAP Overcoming Inaccuracies in Optical Multilayer Perceptrons Moerland, Perry Fiesler, Emile Saxena, Indu activation function liquid crystal light valve (LCLV) non-negative neural networks optical multilayer perceptron weight discretization Proceedings of the First International Symposium on Neuro-Fuzzy Systems (AT'96) Lausanne, Switzerland 1996 AATI All-optical multilayer perceptrons differ in various ways from the ideal neural network model. Examples are the use of non-ideal activation functions which are truncated, asymmetric, and have a non-standard gain, restriction of the network parameters to non-negative values, and the use of limited accuracy for the weights. In this paper an adaptation of the backpropagation learning rule is presented that compensates for these three non-idealities. The good performance of this learning rule is illustrated by a series of experiments. This algorithm enables the implementation of all-optical multilayer perceptrons where learning occurs under control of a computer.