Skip to Main Content
A new self-organizing neural network for the recognition and prediction of multiple-valued patterns is introduced. It is a supervised learning system which incorporates two MVL ART modules (ARTa and ARTb) that can learn to predict a prescribed m-dimensional output vector given a prescribed n-dimensional input vector. These MVL ART modules are linked via an inter ART mapfield that controls the learning of an associative map from ARTa recognition categories to ARTb recognition categories. Facilities are incorporated to predict the correct output vector from learned one-to-many relationships. It can be seen that, learning always converges because all adaptive weights are monotonically nonincreasing. The simulation results show that this network learns for recognition and prediction within four trials while backpropagation requires 100s of trials to perform the same task.