Self-organizing neural networks by dynamic and spatial changing weights

We propose a self-organizing neural structure with dynamic and spatial changing weights for forming a feature space representation of concepts. An essential core of this self-organization is an appropriate combination of an unsupervised learning with incomplete information for the dynamic changing and an extended Hebbian rule for a signal-driven spatial changing. A concept formation problem requires the neural network to acquire the complete feature space structure of concept information using an incomplete observation of the concept. The informational structure can be stored as the connection structure of self-organizing network by using the two rules: the Hebbian rule can create a necessary connection, while unsupervised learning can delete unnecessary connections. Finally concept formation ability of the proposed neural network is proven under some conditions