Bio-synapse behavior controlled by interface engineering in ferroelectric tunnel memristors

Superlow Power Consumption Memristor Based on Borphyrin–Deoxyribonucleic Acid Composite Films as Artificial Synapse for Neuromorphic Computing

“…STDP is a form of the Hebbian learning rule that refers to the time difference (Δt) between the presynaptic and postsynaptic peaks, which greatly influence the strength of the synaptic connection. 50,51 Based on the common asymmetric Hebbian learning rule, if a neuron's prestimulation time precedes its poststimulation time (Δt > 0), the synaptic connection strength can be increased, which indicates long-term potentiation (LTP). Conversely, if the neuron's prestimulation time delays its poststimulation time (Δt < 0), the synaptic connection strength can be decreased, which indicates long-term depression (LTD).…”

“…The synaptic weight refers to the strength of the connection between neurons, and it is changed by the diffusion of Ca 2+ from the synaptic gap to the presynaptic membrane upon arrival of the presynaptic pulse and the release of Ca 2+ from the presynaptic neuron upon receipt of the input pulse. 50 Because the ZnTCPP− DNA-based device is similar in structure to a typical neural synapse and has adjustable conductance, we tested whether it could simulate synaptic plasticity. First, the adjustable conductance was tested by applying 35 positive pulses followed by 35 negative pulses (pulse amplitude of ±4 V, pulse duration, and interval of 1 μs).…”

“…The enlarged view shows the biological synapse between the presynaptic and postsynaptic neurons. The synaptic weight refers to the strength of the connection between neurons, and it is changed by the diffusion of Ca 2+ from the synaptic gap to the presynaptic membrane upon arrival of the presynaptic pulse and the release of Ca 2+ from the presynaptic neuron upon receipt of the input pulse . Because the ZnTCPP–DNA-based device is similar in structure to a typical neural synapse and has adjustable conductance, we tested whether it could simulate synaptic plasticity.…”

Memristors based on NdNiO3 nanocrystals film as sensory neurons for neuromorphic computing

“…[1][2][3][4] A neuromorphic system is constructed of artificial synapses and neurons, and the artificial synapses support intelligent functions such as memorizing and forgetting, learning, and decision-making. 5,6 The neurons begin to accumulate after receiving external stimuli, and output excitatory or inhibitory stimuli to the synapses after reaching a certain threshold a Key Laboratory of Brain-Like Neuromorphic Devices and Systems of Hebei Province, College of Electronic and Information Engineering, Hebei University, Baoding 071002, to realize the adjustment of the weight of the synapses and transmit signals to the next level of neurons in the form of spikes. [7][8][9][10] In order to construct artificial neurons, a variety of artificial neuron models have been proposed based on the memristor, including oscillation neurons, 2,[11][12][13][14][15][16] leaky integratefire (LIF) neurons, [17][18][19][20][21][22][23] Hodgkin-Huxley (H-H) neurons, 24,25 and stochastic neurons.…”

A bidirectional thermal sensory leaky integrate-and-fire (LIF) neuron model based on bipolar NbOx volatile threshold devices with ultra-low operating current

“…10 Nonvolatile memristors are usually used as artificial synapses to realize the storage and computing process of synapse basic functions, such as short-term plasticity (STP), paired-pulse facilitation/depression (PPF/PPD), long-term potentiation/depression (LTP/LTD), consecutive LTP/LTD, and spike-timing-dependent plasticity (STDP). 10–12 To complete the above functions of synapses, synapses need to be stimulated by positive, negative, or bidirectional pulse signals. Therefore, the neurons that can produce bidirectional spikes become a key factor for artificial intelligence systems and a key indicator for the normal operation of an SNN computing system.…”