Fig. 4: 4-neuron circuit for associative memory application.
From: An emergent attractor network in a passive resistive switching circuit
a Attractor map of the 4-neuron network applied with voltages \({{{{\boldsymbol{v}}}}}_{{{{\bf{4}}}}}=\) [0.4; − 1.2; 0.8; 0.8] V. The red dots represent attractor states, and the red arrows indicate the transition paths. In associative memory application, the attractors ‘0001’, ‘0010’, and ‘0011’ are interpreted as cattle, deer, and horses in the daytime, respectively. ‘1001’, ‘1010’, and ‘1011’ are interpreted as the three animals at nighttime. A transition path represents a recalling process of an animal from a given initial vague pattern. b Map of state transition probabilities with consideration of \({V}_{{set}}\), \({V}_{{reset}}\), \({G}_{{HCS}}\) and \({G}_{{LCS}}\) distributions. Every element in the matrix is the probability of an initial state (vertical) transiting to a final state (horizontal). The experimental results are marked by red boxes, which correspond to the transition with the highest probability for most initial states.
