Self-Organizing Maps of Spiking Neurons Using Temporal Coding

Kohonen's self-organizing map has been thoroughly investigated for artificial neural networks. There have been several approaches for biologically more realistic neural networks which all rely on rate coding. Here we show that a topology preserving behavior can be also achieved by networks of spiking neurons using temporal coding. Besides being generally faster during learning and application, our approach has the additional advantage that the winner among competing neurons can be determined fast and locally. Our model is a further step towards a more realistic description of unsupervised learning in biological neural systems. Furthermore, it may provide a basis for fast implementations of neural networks in pulsed VLSI.