초록 ▼

A neural network includes a plurality of input nodes for receiving the respective elements of the input vector. A copy of all of the elements of the input vector is sent to the next level of nodes in the neural network denoted as intermediate nodes. The intermediate nodes each encode a separate temp

A neural network includes a plurality of input nodes for receiving the respective elements of the input vector. A copy of all of the elements of the input vector is sent to the next level of nodes in the neural network denoted as intermediate nodes. The intermediate nodes each encode a separate template pattern. They compare the actual input pattern with the template and generate a signal indicative of the difference between the input pattern and the template pattern. Each of the templates encoded in the intermediate nodes has a class associated with it. The difference calculated by the intermediate nodes is passed to an output node for each of the intermediate nodes at a given class. The output node then selects the minimum difference amongst the values sent from the intermediate nodes. This lowest difference for the class represented by the output node is then forwarded to a selector. The selector receives such values from each of the output nodes of all of the classes and then selects that to output value which is a minimum difference. The selector in turn, generates a signal indicative of the class of the intermediate node that sent the smallest difference value.

대표청구항 ▼

[ I claim:] [1.] A self-organizing neural network for classifying an input pattern signal, which comprises:a) input buffers for receiving scalar element of an input pattern signal vector;b) intermediate nodes, each receiving a copy of said input pattern signal from said input buffers via first signa

[ I claim:] [1.] A self-organizing neural network for classifying an input pattern signal, which comprises:a) input buffers for receiving scalar element of an input pattern signal vector;b) intermediate nodes, each receiving a copy of said input pattern signal from said input buffers via first signal lines, and each generating an output value;c) output nodes, each receiving output values from intermediate nodes of a class via second signal lines, each determining a minimum value among said output values, and each outputting a class signal indicating which intermediate node provided said minimum value and the class signal also indicating the minimum value; andd) a self-organizing selector for determining and outputting both a response signal to an external component indicating the class of the intermediate node which has the minimum value output and a learning signal to said intermediate nodes via third signal lines, wherein said learning signal is calculated from said class signals issued from said output nodes and a teaching signal input from an external component indicating a correct class of said input pattern signal so that weighting factors of said intermediate node are modified based on said learning signal;wherein the self-organizing selector compares the teaching signal to class signals received from the output nodes to determine whether a template of a correct class exists and the learning signal being such that a new intermediate node having weighting factors based on the input pattern and an output node for a new class connected to the intermediate node are created.