TEMPERATURE-CONSTRAINED BACKPROPAGATION NEURAL NETWORKS

A temperature-constrained training algorithm has been devised for backpropagation neural networks. The use of a temperature constraint softens the neural network modeling. This algorithm furnishes more stable models and may decrease training time substantially. These networks are resistant to overfitting and avoid the problem of overtraining. Temperature-constrained and conventional backpropagation networks are evaluated with a synthetic data set and mass spectra from compounds whose molecular formulas are C8H18O. Confusion matrices are used for evaluating classification performance.