Abstract:RRAM-based in-memory computing (IMC) effectively accelerates deep neural networks (DNNs) and other machine learning algorithms. On the other hand, in the presence of RRAM device variations and lower precision, the mapping of DNNs to RRAM-based IMC suffers from severe accuracy loss. In this work, we propose a novel hybrid IMC architecture that integrates an RRAM-based IMC macro with a digital SRAM macro using a programmable shifter to compensate for the RRAM variations and recover the accuracy. The digital SRAM macro consists of a small SRAM memory array and an array of multiply-and-accumulate (MAC) units. The nonideal output from the RRAM macro, due to device and circuit nonidealities, is compensated by adding the precise output from the SRAM macro. In addition, the programmable shifter allows for different scales of compensation by shifting the SRAM macro output relative to the RRAM macro output. On the algorithm side, we develop a framework for the training of DNNs to support the hybrid IMC architecture through ensemble learning. The proposed framework performs quantization (weights and activations), pruning, RRAM IMC-aware training, and employs ensemble learning through different compensation scales by utilizing the programmable shifter. Finally, we design a silicon prototype of the proposed hybrid IMC architecture in the 65-nm SUNY process to demonstrate its efficacy. Experimental evaluation of the hybrid IMC architecture shows that the SRAM compensation allows for a realistic IMC architecture with multilevel RRAM cells (MLCs) even though they suffer from high variations. The hybrid IMC architecture achieves up to 21.9%, 12.65%, and 6.52% improvement in post-mapping accuracy over state-of-the-art techniques, at minimal overhead, for ResNet-20 on CIFAR-10, VGG-16 on CIFAR-10, and ResNet-18 on ImageNet, respectively.

Error Correction Scheme for Reliable RRAM-Based In-Memory Computing

A Circuit-Algorithm Codesign Method to Reduce the Accuracy Drop of RRAM Based Computing-in-memory Chip

Bit-Aware Fault-Tolerant Hybrid Retraining and Remapping Schemes for RRAM-Based Computing-in-Memory Systems

Compensation Architecture to Alleviate Noise Effects in RRAM-based Computing-in-memory Chips with Residual Resource

Intra-array Non-Idealities Modeling and Algorithm Optimization for RRAM-based Computing-in-Memory Applications

A Memristor-Based In-Memory Computing Network for Hamming Code Error Correction

Circuit Modeling for RRAM-Based Neuromorphic Chip Crossbar Array with and Without Write-Verify Scheme.

Accurate Program/Verify Schemes of Resistive Switching Memory (RRAM) for In-Memory Neural Network Circuits

Error Detection and Correction Codes for Safe In-Memory Computations

Compensation Architecture Design Utilizing Residual Resource to Mitigate Impacts of Nonidealities in RRAM-based Computing-in-memory Chips

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

RRAM Based Convolutional Neural Networks for High Accuracy Pattern Recognition and Online Learning Tasks

Online Soft Error Tolerance in ReRAM Crossbars for Deep Learning Accelerators

Rescuing RRAM-Based Computing from Static and Dynamic Faults

A Parallel Multibit Programing Scheme with High Precision for RRAM-Based Neuromorphic Systems

An On-chip Layer-wise Training Method for RRAM Based Computing-in-memory Chips.

A High-Speed and High-Efficiency Diverse Error Margin Write-Verify Scheme for an RRAM-Based Neuromorphic Hardware Accelerator

Theory Study and Implementation of Configurable ECC on RRAM Memory.

Write or Not: Programming Scheme Optimization for RRAM-based Neuromorphic Computing

A 28-Nm RRAM Computing-in-Memory Macro Using Weighted Hybrid 2T1R Cell Array and Reference Subtracting Sense Amplifier for AI Edge Inference

Hybrid RRAM/SRAM in-Memory Computing for Robust DNN Acceleration