In Materia Neuron Spiking Plasticity for Sequential Event Processing Based on Dual‐Mode Memristor

• Published in: Advanced intelligent systems 2022-08, Vol.4 (8), p.n/a
• Main Authors: Shan, Linbo, Wang, Zongwei, Bao, Lin, Bao, Shengyu, Qin, Yabo, Ling, Yaotian, Bai, Guandong, Robertson, John, Cai, Yimao, Huang, Ru
• Format: Article
• Language: English
• Subjects: Adaptation; Algorithms; artificial adaptive neuron; Behavior; Data processing; Energy efficiency; Hafnium oxide; intrinsic plasticity; Memristors; Neural networks; Neuromorphic computing; neuromorphic computing systems; Neurons; Random access memory; spike-rate adaptation; Spiking; Stimuli; Systems stability; Transistors
• ISSN: 2640-4567; 2640-4567
• DOI: 10.1002/aisy.202100264

Artificial neurons are the fundamental elements in neuromorphic computing systems. Studies have revealed neuronal spike‐rate adaptation owing to intrinsic plasticity that neurons will adapt to the spiking patterns and store the events in the background spiking through clustered neuronal spiking. The event can be reactivated by specific retrieval clues instead of solely relying on synaptic plasticity. However, the neural adaptation, as well as the interactive adaptations of neuronal activity for information processing, have not been implemented. Herein, an artificial adaptive neuron via in materia modulation of the VO2/HfO2 based dual‐mode memristor is demonstrated. By changing the conductance of the HfO2 layer, the firing threshold can be modulated, thus the excitability and inhibition can be adjusted according to the previous stimuli without any complex peripherals, showing an adaptive firing rate even under the same stimuli. The artificial neuron clusters can emulate the concept of neuronal memory and neural adaptation, demonstrating spatiotemporal encoding capabilities via the correlated neural firing patterns. This conceptual work provides an alternative way to expand the computation power of spiking neural networks by exploiting the neural adaptation and could be enlightenment to maximize the synergy across both synapse and neuron in neuromorphic computing systems. In materia adaptive neurons is demonstrated based on a dual‐mode VO2/HfO2 memristor. The intrinsic plasticity of neurons can be tuned by modulating the conductance of the HfO2 layer. By encoding letter sequences into temporal inputs, the adaptive neurons successfully process sequential tasks, showing potential to advance spiking neural network (SNN) neuromorphic computing systems.