Dynamics and storage capacity of neural networks with small-world topology

scopus:eid=2-s2.0-78751676189 We study the storage of phase-coded patterns as stable dynamical attractors in recurrent spin neural networks with small-world topology. The synaptic strength of existent connections is determined by a learning rule based on spike-time-dependent plasticity (STDP), with an asymmetric time window depending on the relative timing between pre-and post-synaptic activity. We store multiple patterns and study the network capacity in sparse networks with different topologies. We study networks where each neuron is connected only to a small number z « N of other neurons. Connections can be short range, between neighboring neurons placed on a regular lattice, or long range, between randomly chosen pairs of neurons. We find that a small fraction of long range connections is able to amplify the capacity of the network. This imply that a small-world-network topology maybe optimal, as a compromise between the cost of long range connections and the capacity increase. http://dl.acm.org/citation.cfm?id=1940661&CFID=48978536&CFTOKEN=73330948