Tolerating Noise Effects in Processing-in-Memory Systems for Neural Networks: A Hardware-Software Codesign Perspective

Neural networks have been widely used for advanced tasks from image recognition to natural language processing. Many recent works focus on improving the efficiency of executing neural networks in diverse applications. Researchers have advocated processing-in-memory (PIM) architecture as a promising candidate for training and testing neural networks because PIM design can reduce the communication cost between storage and computing units. However, there exist noises in the PIM system generated from the intrinsic physical properties of both memory devices and the peripheral circuits. The noises introduce challenges in stably training the systems and achieving high test performance, e.g., accuracy in classification tasks. This review discusses the current approaches to tolerating noise effects for both training and inference in PIM systems and provides an analysis from a hardware-software codesign perspective. Noise-tolerant strategies for PIM systems based on resistive random-access memory (ReRAM), including circuit-level, algorithm-level, and system-level solutions are explained. In addition, we also present some selected noise-tolerate cases in PIM systems for generative adversarial networks and physical neural networks.

Automated summary

This review discusses the current approaches to tolerating noise effects in processing-in-memory (PIM) systems. PIM architecture is considered a promising solution for reducing communication cost between storage and computing units, but noises generated from memory and peripheral circuits pose challenges. The review explains noise-tolerant strategies for PIM systems based on resistive random-access memory (ReRAM) and presents case studies for generative adversarial networks and physical neural networks.