Digital design of sigmoid approximator for artificial neural networks

Electronics Letters

Volumn 38, Issue 1, 2002, Pages 35-37

  Basterretxea, K ^a   Tarela, J M ^a   Del Campo, I ^a  

^a UNIVERSITY OF THE BASQUE COUNTRY UPV EHU   (Spain)

Author keywords

[No Author keywords available]

Indexed keywords

APPROXIMATION THEORY; FUNCTION EVALUATION; INTERPOLATION; LEARNING ALGORITHMS; NEURAL NETWORKS; PIECEWISE LINEAR TECHNIQUES; RECURSIVE FUNCTIONS;

CENTRED RECURSIVE INTERPOLATION; PIECEWISE LINEAR APPROXIMATION; SIGMOID FUNCTION;

VLSI CIRCUITS;

EID: 0037012092     PISSN: 00135194     EISSN: None     Source Type: Journal    
DOI: 10.1049/el:20020008     Document Type: Article

References (6)

1. Efficient implementation of piecewise linear activation function for digital VLSI neural networks
2. Simple approximation of sigmoidal functions: Realistic design of digital neural networks capable of learning
3. Piecewise linear approximation applied to nonlinear function of a neural network
4. Implementation issues of sigmoid function and its derivative for VLSI digital neural networks
5. Optimised PWL recursive approximation and its application to neuro-fuzzy systems
6. Sigmoid generators for neural computing using piecewise approximations