Ritter H
Dept. of Phys., Illinois Univ., Urbana, IL.
IEEE Trans Neural Netw. 1991;2(1):173-5. doi: 10.1109/72.80310.
It is shown that for a class of vector quantization processes, related to neural modeling, that the asymptotic density Q(x ) of the quantization levels in one dimension in terms of the input signal distribution P(x) is a power law Q(x)=C-P(x)(alpha ), where the exponent alpha depends on the number n of neighbors on each side of a unit and is given by alpha=2/3-1/(3n (2)+3n+1). The asymptotic level density is calculated, and Monte Carlo simulations are presented.
结果表明，对于一类与神经建模相关的矢量量化过程，就输入信号分布P(x)而言，一维量化电平的渐近密度Q(x)是一个幂律Q(x)=C·P(x)^(α)，其中指数α取决于单位每一侧的邻居数n，且由α = 2/3 - 1/(3n² + 3[n + 1]²)给出。计算了渐近电平密度，并给出了蒙特卡罗模拟结果。
