Robust pattern recognition using non-iteratively learned perceptron

Whenever the input training patterns applied to a one-layered, hard-limited perceptron (OHP) satisfy a certain positive-linear-independency (PLI) condition, the learning of these standard patterns by the neural network can be done non-iteratively in a few algebraic steps and the recognition of the untrained test patterns can reach an ''optimal robustness'' if a special learning scheme is adopted in the learning mode. In this paper, we report the theoretical foundation, the analysis (design) of this pattern recognition system, and the experiments we carried out with this novel system. The experimental result shows that the learning of four digitized training patterns is close to real-time, and the recognition of the untrained patterns is above 90% correct. The ultra-fast learning speed we achieved here is due to the non-iterative nature of the novel learning scheme. The high robustness in recognition here is due to the optimal robustness analysis (including a special feature extraction process) we used in the neural network design.