A neural network method for reducing data dimensionality based on the concept of input training, in which each input pattern is not fired but adjusted along with internal network parameters to reproduce its corresponding output pattern, is presented. With input adjustment, a properly configured network can be trained to reproduce a given data set with minimum distortion; the trained network inputs provide reduced data. A three-layer network with input training can perform all functions of a five-layer autoassociative network, essentially capturing nonlinear correlations among data. In addition, simultaneous training of a network and its inputs is shown to be significantly more efficient in reducing data dimensionality than training an autoassociative network. The concept of input training is closely related to principal component analysis (PCA) and the principal curve method, which is a nonlinear extension of PCA.