Constructive feedforward neural networks using Hermite polynomial activation functions

Authors:
Liying Ma;K. Khorasani
Affiliations:
Dept. of Electr. & Comput. Eng., Concordia Univ., Montreal, Que., Canada;-
Venue:
IEEE Transactions on Neural Networks
Year:
2005

Citing 0
Cited 18

Neural Structures Using the Eigenstates of a Quantum Harmonic Oscillator

Open Systems & Information Dynamics
A study on the modeling ability of the IDS method: A soft computing technique using pattern-based information processing

International Journal of Approximate Reasoning
Training neural networks for classification using growth probability-based evolution

Neurocomputing
Properties of the Hermite Activation Functions in a Neural Approximation Scheme

ICANNGA '07 Proceedings of the 8th international conference on Adaptive and Natural Computing Algorithms, Part II
A New Constructive Algorithm for Designing and Training Artificial Neural Networks

Neural Information Processing
Several Enhancements to Hermite-Based Approximation of One-Variable Functions

ICANN '08 Proceedings of the 18th international conference on Artificial Neural Networks, Part I
Image classification of artificial fingerprints using Gabor wavelet filters, self-organising maps and Hermite/Laguerre neural networks

International Journal of Knowledge Engineering and Soft Data Paradigms
A new adaptive merging and growing algorithm for designing artificial neural networks

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
A novel pruning algorithm for self-organizing neural network

IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Neural networks with asymmetric activation function for function approximation

IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
A new constructive algorithm for architectural and functional adaptation of artificial neural networks

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Some enhancements to orthonormal approximation of 2D functions

ICANNGA'09 Proceedings of the 9th international conference on Adaptive and natural computing algorithms
Approximation of functions by multivariable hermite basis: a hybrid method

ICANNGA'11 Proceedings of the 10th international conference on Adaptive and natural computing algorithms - Volume Part I
Feed-Forward neural networks using hermite polynomial activation functions

SETN'06 Proceedings of the 4th Helenic conference on Advances in Artificial Intelligence
A method of multivariable Hermite basis function approximation

Neurocomputing
Adaptive PI Hermite neural control for MIMO uncertain nonlinear systems

Applied Soft Computing
Efficient self-organizing multilayer neural network for nonlinear system modeling

Neural Networks
Optimization of the weights and asymmetric activation function family of neural network for time series forecasting

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a constructive one-hidden-layer network is introduced where each hidden unit employs a polynomial function for its activation function that is different from other units. Specifically, both a structure level as well as a function level adaptation methodologies are utilized in constructing the network. The functional level adaptation scheme ensures that the "growing" or constructive network has different activation functions for each neuron such that the network may be able to capture the underlying input-output map more effectively. The activation functions considered consist of orthonormal Hermite polynomials. It is shown through extensive simulations that the proposed network yields improved performance when compared to networks having identical sigmoidal activation functions.