Abstract:Spiking neural networks (SNNs) are biologically plausible and computationally powerful. The current computing systems based on the von Neumann architecture are almost the hardware basis for the implementation of SNNs. However, performance bottlenecks in computing speed, cost, and energy consumption hinder the hardware development of SNNs. Therefore, efficient non von Neumann hardware computing systems for SNNs remain to be explored. In this article, a selective supervised algorithm for spiking neurons (SNs) inspired by the selective attention mechanism is proposed, and a memristive SN circuit as well as a memristive SNN circuit based on the proposed algorithm are designed. The memristor realizes the learning and memory of the synaptic weight. The proposed algorithm includes a top-down (TD) selective supervision method and a bottom-up (BU) selective supervision method. Compared with other supervised algorithms, the proposed algorithm has excellent performance on sequence learning. Moreover, TD and BU attention encoding circuits are designed to provide the hardware foundation for encoding external stimuli into TD and BU attention spikes, respectively. The proposed memristive SNN circuit can perform classification on the MNIST dataset and the Fashion-MNIST dataset with superior accuracy after learning a small number of labeled samples, which greatly reduces the cost of manual annotation and improves the supervised learning efficiency of the memristive SNN circuit.

Saturation Region Attractivity Analysis for Memristive Neural Networks under the Strong External Input

Stability Analysis for Memristive Recurrent Neural Network Under Different External Stimulus

Attractor flow analysis for recurrent neural network with back-to-back memristors

Associative Memory Synthesis Based on Region Attractive Recurrent Neural Networks

Attractivity Analysis of Memristor-Based Cellular Neural Networks with Time-Varying Delays

Memory Analysis for Memristors and Memristive Recurrent Neural Networks.

Convergence and Attractivity of Memristor-Based Cellular Neural Networks with Time Delays

Multistability of periodic delayed recurrent neural network with memristors

Region stability analysis and tracking control of memristive recurrent neural network

Analysis and Design of Associative Memories Based on Recurrent Neural Networks with Linear Saturation Activation Functions and Time-Varying Delays

H∞ Control Design for Memristor-Based Neural Networks Subject to Actuator Saturation

An improved criterion for stability and attractability of memristive neural networks with time-varying delays.

A Memristive Spiking Neural Network Circuit with Selective Supervised Attention Algorithm

Overview of Memristor-Based Neural Network Design and Applications

Memristor Neural Network Design

High Robustness Memristor Neural State Machines

A Multi-Stable Memristor and its Application in a Neural Network

Dynamic Behaviors of Hybrid Lotka-Volterra Recurrent Neural Networks with Memristor Characteristics.

Symmetrical Coexisting Attractors and Extreme Multistability Induced by Memristor Operating Configurations in SC-CNN

A novel memristive synapse-coupled ring neural network with countless attractors and its application

Global Attractivity of Discrete-Time Recurrent Neural Networks with Lnsaturating Piecewise Linear Activation Functions