Contraction Properties of VLSI Cooperative Competitive Neural Networks of Spiking Neurons

A non–linear dynamic system is called contracting if initial conditions are forgotten exponentially fast, so that all trajectories converge to a single trajectory. We use contraction theory to derive an upper bound for the strength of recurrent connections that guarantees contraction for complex neural networks. Specifically, we apply this theory to a special class of recurrent networks, often called ive Competitive Networks (CCNs), which are an abstract representation of the cooperative-competitive connectivity observed in cortex. This specific type of network is believed to play a major role in shaping cortical responses and selecting the relevant signal among distractors and noise. In this paper, we analyze contraction of combined CCNs of linear threshold units and verify the results of our analysis in a hybrid analog/digital VLSI CCN comprising spiking neurons and dynamic synapses.

• Type: Conference
• Year: 2007
• Where: NIPS
• Authors: Emre Neftci, Elisabetta Chicca, Giacomo Indiveri, Jean-Jacques E. Slotine, Rodney J. Douglas
• Keywords: Complex Neural Networks, Contraction Theory, Information Technology, Ive Competitive Networks, NIPS 2007