A neural network circuit with associative learning and forgetting process based on memristor neuromorphic device

Memristor has attracted extensive attention in recent years because of its capability to act as artificial synapse in neuromorphic network. In this paper, by combining a designed memristive synapse and a resistive synapse, a novel neural network circuit with three neurons and one operational module has been constructed. With the memristor device whose resistance can be purposely changed according to the time interval between pre- and post-spikes, the weight of the memristive synapse that connected between two neurons can be modified as required. What's more, the transmission of signal from pre-neuron to post-neuron and the adjustment of memristive resistance were carried out alternately through the coordination of different modules. Finally, the neural network simulated the Pavlov's experiment successfully, indicating that it can realize the process of associative learning and forgetting.