Abstract:Brain-inspired neuromorphic computing has emerged as a promising solution to overcome the energy and speed limitations of conventional von Neumann architectures. In this context, in-memory computing utilizing memristors has gained attention as a key technology, harnessing their non-volatile characteristics to replicate synaptic behavior akin to the human brain. However, challenges arise from non-linearities, asymmetries, and device variations in memristive devices during synaptic weight updates, leading to inaccurate weight adjustments and diminished recognition accuracy. Moreover, the repetitive weight updates pose endurance challenges for these devices, adversely affecting latency and energy consumption. To address these issues, we propose a Siamese network learning approach to optimize the training of multi-level memristor neural networks. During neural inference, forward propagation takes place within the memristor neural network, enabling error and noise detection in the memristive devices and hardware circuits. Simultaneously, high-precision gradient computation occurs on the software side, initially updating the floating-point weights within the Siamese network with gradients. Subsequently, weight quantization is performed, and the memristor conductance values requiring updates are modified using a sparse update strategy. Additionally, we introduce gradient accumulation and weight quantization error compensation to further enhance network performance. The experimental results of MNIST data recognition, whether based on a MLP or a CNN model, demonstrate the rapid convergence of our network model. Moreover, our method successfully eliminates over 98% of weight updates for memristor conductance weights within a single epoch. This substantial reduction in weight updates leads to a significant decrease in energy consumption and time delay by more than 98% when compared to the basic closed-loop update method. Consequently, this approach effectively addresses the durability requirements of memristive devices.

A Multi-conductance States Memristor-based CNN Circuit Using Quantization Method for Digital Recognition.

Optimized Near-Zero Quantization Method for Flexible Memristor Based Neural Network

Pruning and quantization algorithm with applications in memristor-based convolutional neural network

Fully hardware-implemented memristor convolutional neural network

Hybrid Dual-Complementary Metal-Oxide-semiconductor/memristor Synapse-Based Neural Network with Its Applications in Image Super-Resolution

Memristor-Based Cellular Nonlinear/Neural Network: Design, Analysis, and Applications

Enhancing in-situ updates of quantized memristor neural networks: a Siamese network learning approach

Memristor-Based Circuit Design for Multilayer Neural Networks

Memristor-based in-situ convolutional strategy for accurate braille recognition

Memristor crossbar deep network implementation based on a Convolutional neural network

Study of Weight Quantization Associations over a Weight Range for Application in Memristor Devices

QuantBayes: Weight Optimization for Memristive Neural Networks via Quantization-Aware Bayesian Inference

In Situ Training of Feed-Forward and Recurrent Convolutional Memristor Networks

Convolutional Neural Networks Based on RRAM Devices for Image Recognition and Online Learning Tasks

Memristor-based Deep Convolution Neural Network: A Case Study

Implementation of Convolutional Neural Networks in Memristor Crossbar Arrays with Binary Activation and Weight Quantization

Associative learning of a three-terminal memristor network for digits recognition

Implementation of Multiple-Step Quantized STDP Based on Novel Memristive Synapses

An Ultra-Efficient Memristor-Based DNN Framework with Structured Weight Pruning and Quantization Using ADMM

Some Interesting Features of Memristor CNN

Dual Redox-active Covalent Organic Framework-based Memristors for Highly-efficient Neuromorphic Computing