Vortex

TLDR

Recent advances in memristor devices and crossbar integration enable low‑power on‑chip neuromorphic computing, yet the robustness of training methods under realistic hardware constraints remains largely unexplored. This study quantitatively evaluates how device imperfections and circuit design constraints affect the robustness of close‑loop on‑device and open‑loop off‑device training, and introduces Vortex, a variation‑aware scheme to enhance robustness. The authors analyze the effects of device variations and circuit constraints on the two training methods and develop Vortex, which compensates for variations and optimizes crossbar mapping. Vortex improves test accuracy by 29.6 % over OLD and 26.4 % over CLD on average.

Abstract

Recent advances in development of memristor devices and crossbar integration allow us to implement a low-power on-chip neuromorphic computing system (NCS) with small footprint. Training methods have been proposed to program the memristors in a crossbar by following existing training algorithms in neural network models. However, the robustness of these training methods has not been well investigated by taking into account the limits imposed by realistic hardware implementations. In this work, we present a quantitative analysis on the impact of device imperfections and circuit design constraints on the robustness of two popular training methods -- "close-loop on-device" (CLD) and "open-loop off-device" (OLD). A novel variation-aware training scheme, namely, Vortex, is then invented to enhance the training robustness of memristor crossbar-based NCS by actively compensating the impact of device variations and optimizing the mapping scheme from computations to crossbars. On average, Vortex can significantly improve the test rate by 29.6% and 26.4%, compared to the traditional OLD and CLD, respectively.