Designing a Pattern Recognition Neural Network with a Reject Output and Many Sets of Weights and Biases

Abstract

Adding the reject output to the pattern recognition neural network is an approach to help the neural network can classify almost all patterns of a training data set by using many sets of weights and biases, even if the neural network is small. With a smaller number of neurons, we can implement the neural network on a hardware-based platform more easily and also reduce the response time of it. With the reject output the neural network can produce not only right or wrong results but also reject results. It is significant, if we design a neural network to help a robot to interact with people. The reject results can be accepted by the robot in this interaction process. If the neural network rejected a pattern, the robot would ask people to make the pattern again that looks like we talk “Pardon me”.