Unsupervised Classification of Complex Clusters in Networks of Spiking Neurons

For unsupervised clustering in a network of spiking neurons we develop a temporal encoding of continuously valued data to obtain arbitrary clustering capacity and precision with an efficient use of neurons. Input variables are encoded independently in a population code by neurons with 1-dimensional graded and overlapping sensitivity profiles. Using a temporal Hebbian learning rule, the network architecture yields reliable clustering of high-dimensional multi-modal data. Additionally, multi-scale sensitivity to the input is achieved by using an appropriate choice of local activation functions. To correctly classify non-spherical clusters, we present a multi-layer version of the algorithm to perform a form of hierarchical clustering. Together with the addition of lateral excitatory connections in the hidden layer, this enables the correct classification of complex, interlocking clusters by synchronizing the neurons coding for parts of the same cluster. We show how synchronous spiking of...

• Type: Conference
• Year: 2000
• Where: IJCNN
• Authors: Sander M. Bohte, Johannes A. La Poutré, Joost N. Kok
• Keywords: Arbitrary Clustering Capacity, Hebbian Learning Rule, IJCNN 2000, Neural Networks, Neurons