Although synaptic behaviours of memristors have been widely demonstrated, implementation of an even simple artificial neural network is still a great challenge. In this work, we demonstrate the associative memory on the basis of a memristive Hopfield network. Different patterns can be stored into the memristive Hopfield network by tuning the resistance of the memristors, and the pre-stored patterns can be successfully retrieved directly or through some associative intermediate states, being analogous to the associative memory behaviour. Both single-associative memory and multi-associative memories can be realized with the memristive Hopfield network.
Synaptic Long-Term Potentiation (LTP), which is a long-lasting enhancement in signal transmission between neurons, is widely considered as the major cellular mechanism during learning and memorization. In this work, a NiOx-based memristor is found to be able to emulate the synaptic LTP. Electrical conductance of the memristor is increased by electrical pulse stimulation and then spontaneously decays towards its initial state, which resembles the synaptic LTP. The lasting time of the LTP in the memristor can be estimated with the relaxation equation, which well describes the conductance decay behavior. The LTP effect of the memristor has a dependence on the stimulation parameters, including pulse height, width, interval, and number of pulses. An artificial network consisting of three neurons and two synapses is constructed to demonstrate the associative learning and LTP behavior in extinction of association in Pavlov's dog experiment.
Although there is a huge progress in complementary-metal-oxide-semiconductor (CMOS) technology, construction of an artificial neural network using CMOS technology to realize the functionality comparable with that of human cerebral cortex containing 1010–1011 neurons is still of great challenge. Recently, phase change memristor neuron has been proposed to realize a human-brain level neural network operating at a high speed while consuming a small amount of power and having a high integration density. Although memristor neuron can be scaled down to nanometer, integration of 1010–1011 neurons still faces many problems in circuit complexity, chip area, power consumption, etc. In this work, we propose a CMOS compatible HfO2 memristor neuron that can be well integrated with silicon circuits. A hybrid Convolutional Neural Network (CNN) based on the HfO2 memristor neuron is proposed and constructed. In the hybrid CNN, one memristive neuron can behave as multiple physical neurons based on the Time Division Multiplexing Access (TDMA) technique. Handwritten digit recognition is demonstrated in the hybrid CNN with a memristive neuron acting as 784 physical neurons. This work paves the way towards substantially shrinking the amount of neurons required in hardware and realization of more complex or even human cerebral cortex level memristive neural networks.
Ge nanocrystals ͑nc-Ge͒ embedded in the gate oxide of the nonvolatile memory structure were synthesized by Ge ion implantation followed by thermal annealing at 800°C for various durations. Large changes in the structural and chemical properties of the Ge + -implanted oxide have been observed, and they have been found to possess a significant impact on the charge transfer in the oxide layer. The distribution and concentration of the nc-Ge and dissolved Ge atoms which serve as both the charge storage and transfer sites in the oxide are affected by the annealing. Two charge transfer mechanisms, i.e., the lateral charge diffusion along the Ge-distributed layer and the charge leakage from the charge storage sites to the Si substrate via the charge transfer sites, have been identified based on the charge retention behaviors. Both mechanisms are enhanced by the annealing as a result of the change in the distribution and concentration of the charge transfer sites.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.