Stochastic vs. Deterministic Neural Networks for Pattern Recognition

Abstract

The performance of several neural network-like models for pattern recognition tasks are analyzed. A comparison based on recognition of random points in a multidimensional space is made among Backpropagation and different variations of Learning Vector Quantization and Boltzmann machine. The Boltzmann machine models and Hierarchical Learning Vector Quantization are found to perform well in the investigated tasks.