SIGNIFICANCE OF DISTRIBUTED REPRESENTATION IN THE OUTPUT LAYER OF A NEURAL-NETWORK IN A PATTERN-RECOGNITION TASK

In the cerebral cortex, it is assumed that information is represented by the activity pattern of an assembly of neurons and the synaptic eff icacies among then. A distributed representation of pattern is incorpo rated in the output layer of a neural network with an error back-propa gation algorithm, in order to study its technological merits. The netw ork has three layers, which consist of a 32 x 32 array of units (1024) for the input layer, 6-25 units for the hidden layer and 12 units for the output layer. 12 triangular patterns with a variety of parameters are presented to the input layer. Three output-layer units are assign ed to each input figure. After initial learning, the network responds to the learned patterns, showing a generalisation for patterns. The ne twork shows resistance to unit de-activation procedures. When the inpu t layer is exposed to the learned pattern, the hidden-layer units show associative activation pattern. These results indicate that the organ isation of information representation in the output layer in a neural network strongly influences both the performance of the whole network and information representation in the hidden layer.