AN ANALYSIS OF THE GLVQ ALGORITHM

Generalized learning vector quantization (GLVQ) has bees proposed in [5] as a generalization of the simple competitive learning (SCL) algorithm. The main argument of GLVQ proposal is its superior insensitivity to the initial values of the weights (codevectors). In this letter we show that the distinctive characteristics of the definition of GLVQ disappear outside a small domain of applications. GLVQ becomes identical to SCL when either the number of codevectors grows or the size of the input space is large. Besides that, the behavior of GLVQ is inconsistent for problems defined on very small scale input spaces. The adaptation rules fluctuate between performing descent and ascent searches on the gradient of the distortion function.