Single-layer Perception with 4x4 input neurons and one output neuron is optically implemented. Holographic lenslet arrays are used for the programmable optical interconnection topology. The hologram is bleached in order to increase the diffraction efficiency.
It is shown that the performance of Perception depends on the learning rate, the inertia rate, and the correlation of input patterns. To shown the programmability of our interconnection scheme, another learning network(a dynamic Hopfield-like neural network) is optically implemented in the appendix.