Supervised Neural Gas with General Similarity Measure
作者:Barbara Hammer, Marc Strickert, Thomas Villmann
摘要
Prototype based classification offers intuitive and sparse models with excellent generalization ability. However, these models usually crucially depend on the underlying Euclidian metric; moreover, online variants likely suffer from the problem of local optima. We here propose a generalization of learning vector quantization with three additional features: (I) it directly integrates neighborhood cooperation, hence is less affected by local optima; (II) the method can be combined with any differentiable similarity measure whereby metric parameters such as relevance factors of the input dimensions can automatically be adapted according to the given data; (III) it obeys a gradient dynamics hence shows very robust behavior, and the chosen objective is related to margin optimization.
论文关键词:generalized relevance LVQ, learning vector quantization, metric adaptation, neural gas
论文评审过程:
论文官网地址:https://doi.org/10.1007/s11063-004-3255-2