Performance evaluation of a "parallel collision control" unsupervised neural network
Giuseppe Acciani, E. Chiarantoni, Maria Teresa Minenna, Francesco Vacca · 2002
In this paper a new neural approach to clustering tasks in handwritten numeral recognition problems is compared to classical unsupervised neural networks techniques. The kernel of the proposed network is a neural unit able to perform clustering acting alone. The network is able to find directly dense zones of the input space without requiring competition and thus overcoming the major drain backs of classical unsupervised architectures.