Similarities of error regularization, sigmoid gain scaling, target smoothing, and training with jitter

Authors:
R. Reed;R. J. Marks, II;S. Oh
Affiliations:
Dept. of Electr. Eng., Washington Univ., Seattle, WA;-;-
Venue:
IEEE Transactions on Neural Networks
Year:
1995

Citing 0
Cited 14

Noise injection: theoretical prospects

Neural Computation
Classifier's Complexity Control while Training Multilayer Perceptrons

Proceedings of the Joint IAPR International Workshops on Advances in Pattern Recognition
Using the Hermite Regression Formula to Design a Neural Architecture with Automatic Learning of the "Hidden" Activation Functions

AI*IA '99 Proceedings of the 6th Congress of the Italian Association for Artificial Intelligence on Advances in Artificial Intelligence
On different facets of regularization theory

Neural Computation
A neural network ensemble method with jittered training data for time series forecasting

Information Sciences: an International Journal
Improving performance of neural classifiers via selective reduction of target levels

Neurocomputing
Designing Model Based Classifiers by Emphasizing Soft Targets

Fundamenta Informaticae - Advances in Artificial Intelligence and Applications
Convergence and objective functions of some fault/noise-injection-based online learning algorithms for RBF networks

IEEE Transactions on Neural Networks
Comparison of evolutionary computation techniques for noise injected neural network training to estimate longitudinal dispersion coefficients in rivers

Expert Systems with Applications: An International Journal
Mining data by query-based error-propagation

ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part I
Disease diagnosis using query-based neural networks

ISNN'05 Proceedings of the Second international conference on Advances in Neural Networks - Volume Part III
Establishing criteria to ensure successful feedforward artificial neural network modelling of mechanical systems

Mathematical and Computer Modelling: An International Journal
Approximation properties of local bases assembled from neural network transfer functions

Mathematical and Computer Modelling: An International Journal
Designing Model Based Classifiers by Emphasizing Soft Targets

Fundamenta Informaticae - Advances in Artificial Intelligence and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

The generalization performance of feedforward layered perceptrons can, in many cases, be improved either by smoothing the target via convolution, regularizing the training error with a smoothing constraint, decreasing the gain (i.e., slope) of the sigmoid nonlinearities, or adding noise (i.e., jitter) to the input training data, In certain important cases, the results of these procedures yield highly similar results although at different costs. Training with jitter, for example, requires significantly more computation than sigmoid scaling