The principle of weight divergence facilitation for unsupervised pattern recognition in spiking neural networks

Parallels between the signal processing tasks and biological neurons lead to\nan understanding of the principles of self-organized optimization of input\nsignal recognition. In the present paper, we discuss such similarities among\nbiological and technical systems. We propose adding the well-known STDP\nsynaptic plasticity rule to direct the weight modification towards the state\nassociated with the maximal difference between background noise and correlated\nsignals. We use the principle of physically constrained weight growth as a\nbasis for such weights' modification control. It is proposed that the existence\nand production of bio-chemical 'substances' needed for plasticity development\nrestrict a biological synaptic straight modification. In this paper, the\ninformation about the noise-to-signal ratio controls such a substances'\nproduction and storage and drives the neuron's synaptic pressures towards the\nstate with the best signal-to-noise ratio. We consider several experiments with\ndifferent input signal regimes to understand the functioning of the proposed\napproach.\n