Abstract-This paper presents the tuning of the structure and parameters of a neural network using an improved genetic algorithm (GA). It will also be shown that the improved GA performs better than the standard GA based on some benchmark test functions. A neural network with switches introduced to its links is proposed. By doing this, the proposed neural network can learn both the input-output relationships of an application and the network structure using the improved GA. The number of hidden nodes is chosen manually by increasing it from a small number until the learning performance in terms of fitness value is good enough. Application examples on sunspot forecasting and associative memory are given to show the merits of the improved GA and the proposed neural network.Index Terms-Genetic algorithm (GA), neural networks, parameter learning, structure learning.

No abstract

Patients with idiopathic thrombocytopenic purpura who have good or excellent responses to intravenous immune globulin are likely to have good or excellent responses to splenectomy, whereas patients who have poor responses to intravenous immune globulin are unlikely to have good or excellent responses to splenectomy.

A chaotic filter bank for computer cryptography is proposed. By encrypting and decrypting signals via a chaotic filter bank, the following advantages are enjoyed: 1) one can embed signals in different frequency bands by employing different chaotic functions; 2) the number of chaotic generators to be employed and their corresponding functions can be selected and designed in a flexible manner because perfect reconstruction does not depend on the invertibility, causality, linearity and time invariance of the corresponding chaotic functions; 3) the ratios of the subband signal powers to the chaotic subband signal powers can be easily changed by the designers and perfect reconstruction is still guaranteed no matter how small these ratios are; 4) the proposed cryptographical system can be easily adapted in the international multimedia standards, such as JPEG 2000 and MPEG4.