Fig. 6: Controlling network dynamical states with triangular AC signals.
From: Avalanches and edge-of-chaos learning in neuromorphic nanowire networks
a For a 100 nanowire, 261 junction simulated network, the driving amplitude A and frequency f controls the maximal Lyapunov exponent λ and hence, dynamical evolution of the network: λ < 0 indicates convergence to stable attractor states; λ > 0 indicates chaotic dynamics; and λ ≈ 0 corresponds to an edge-of-chaos state. b For a given λ, the collective network response is captured by the average ratio of maximum-to-minimum network conductance states, r.
