遗传选择机制在矢量量化中的应用

提出误差选择竞争学习算法，它把遗传算法中的选择机制引入到矢量量化设计中，在使用竞争学习算法减小期望误差的前提下，利用选择机制调整各个区域的子误差从而进一步改善期望误差，实验结果表明，该算法较好地调整了各区域的子误差，克服局部最优

Application of Genetic Selection Mechanism for Vector Quantization

An learning algorithm for distortion selection competitione is presented. It introduces the selection mechanism of genetic algorithm into vector quantization. After competitive learning (CL) algorithm has been used to decrease the expected distortion, the selection mechanism is utilized to modulate the subdistortion of each region in order to improve the expected distortion. The experimental results show that the algorithm has better modulated the subdistortion of each region and overcome the local optimality.