Analysis of the Initial Values in Split-Complex Backpropagation Algorithm

• Published in: IEEE transactions on neural networks Vol. 19; no. 9; pp. 1564 - 1573
• Main Authors: Sheng-Sung Yang, Siu, S., Chia-Lu Ho
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.09.2008; Institute of Electrical and Electronics Engineers
• Subjects: Adjustment; Algorithm design and analysis; Algorithms; Applied sciences; Artificial Intelligence; Back propagation; Backpropagation (BP); Backpropagation algorithms; Channels; Communication channels; Computer science; control theory; systems; Computer Simulation; Connectionism. Neural networks; Convergence; Cost function; Criteria; Decision feedback equalizers; Exact sciences and technology; initial values; Models, Theoretical; multilayer perceptron (MLP); Multilayer perceptrons; Network topology; Neural networks; Neural Networks (Computer); Pattern Recognition, Automated - methods; Signal processing algorithms; Weight reduction; Backpropagation; multilayer perceptron (MLP); Backpropagation algorithm; Backpropagation (BP); initial values; Initial value problem; Multilayer perceptrons; Equalizer; Neural network; Equalization; Multilayer network
• ISSN: 1045-9227; 1941-0093; 1941-0093
• DOI: 10.1109/TNN.2008.2000805

When a multilayer perceptron (MLP) is trained with the split-complex backpropagation (SCBP) algorithm, one observes a relatively strong dependence of the performance on the initial values. For the effective adjustments of the weights and biases in SCBP, we propose that the range of the initial values should be greater than that of the adjustment quantities. This criterion can reduce the misadjustment of the weights and biases. Based on the this criterion, the suitable range of the initial values can be estimated. The results show that the suitable range of the initial values depends on the property of the used communication channel and the structure of the MLP (the number of layers and the number of nodes in each layer). The results are studied using the equalizer scenarios. The simulation results show that the estimated range of the initial values gives significantly improved performance.