Fig. 2: Threshold network activation under constant bias.
From: Avalanches and edge-of-chaos learning in neuromorphic nanowire networks
Simulations using a 100 nanowire, 261 junction NWN. a DC activation curves for an initially inactive NWN (Λ = 0 for all junctions) for increasing applied voltages (V* = V/Vth), with Vth = 0.09 V. b Snapshot visualisation of NWN, showing formation of first transport pathway, corresponding to shortest path length n. c Steady-state network conductance (with \({G}_{{{\rm{nw}}}}^{* }=n{G}_{nw}/{{{\rm{G}}}}_{0}\)), as a function of V*. Black circles are for an initially inactive network. Red triangles are for a network pre-activated by a 1.8 V DC bias. Inset shows a zoom-in near V* = 1 with a logarithmic vertical scale.
