Neural network for regression problems with reduced training sets

Abstract

Although they are powerful and successful in many applications, artificial neural networks (ANNs) typically do not perform well with complex problems that have a limited number of training cases. Often, collecting additional training data may not be feasible or may be costly. Thus, this work presents a new radial-basis network (RBN) design that overcomes the limitations of using ANNs to accurately model regression problems with minimal training data. This new design involves a multi-stage training process that couples an orthogonal least squares (OLS) technique with gradient-based optimization. New termination criteria are also introduced to improve accuracy. In addition, the algorithms are designed to require minimal heuristic parameters, thus improving ease of use and consistency in performance. The proposed approach is tested with experimental and practical regression problems, and the results are compared with those from typical network models. The results show that the new design demonstrates improved accuracy with reduced dependence on the amount of training data. As demonstrated, this new ANN provides a platform for approximating potentially slow but high-fidelity computational models, and thus fostering inter-model connectivity and multi-scale modeling.

The use of urban indicators in forecasting a real estate value with the use of deep neural network

“…If the network's answer is not satisfactory (mistake is too big) a modi cation of the network's weights take place, in such a way that at the next presentation of the example, the error is smaller. The majority of the neural network training algorithms belong to gradient methods of training (Bataineh and Marler, 2017). However, their detailed explanation exceeds the capabilities of this work.…”

Parametric Shape Modeling and Skeleton Extraction With Radial Basis Functions Using Similarity Domains Network

“…Their approach utilizes first computing the standard deviation of each cluster (after applying a k-means like clustering on the data) and then using a scaled version of those standard deviations of each cluster as the shape parameter for each RBF in the network. The work in [2] also used a similar approach by using the root-meansquare-deviation (RMSD) value between the RBF centers and the data value for each RBF in the network. The authors used a modified orthogonal least squares (OLS) algorithm to select the RBF centers.…”

Developing New Solutions for a Reconfigurable Microstrip Patch Antenna by Inverse Artificial Neural Networks

“…where is the number of training data samples and is the number of frequency samples which is equal to 200. The large amount of training data has been reduced to be 27 samples only (Bataineh and Marler, 2017). The reduction procedure depends on the selection of resonant frequency samples from the training data.…”