Processing-In-Memory Acceleration of Convolutional Neural Networks for Energy-Effciency, and Power-Intermittency Resilience

Herein, a bit-wise Convolutional Neural Network (CNN) in-memory accelerator is implemented using Spin-Orbit Torque Magnetic Random Access Memory (SOT-MRAM) computational sub-arrays. It utilizes a novel AND-Accumulation method capable of significantly-reduced energy consumption within convolutional layers and performs various low bitwidth CNN inference operations entirely within MRAM. Power-intermittence resiliency is also enhanced by retaining the partial state information needed to maintain computational forward-progress, which is advantageous for battery-less IoT nodes. Simulation results indicate $\sim \pmb{5.4}\times\mathbf{ higher}$ energy-efficiency and $9\times\mathbf{speedup}$ over ReRAM-based acceleration, or roughly $\sim \pmb{9.7}\times \mathbf{higher}$ energy-efficiency and $\pmb{13.5}\times\mathbf{speedup}$ over recent CMOS-only approaches, while maintaining inference accuracy comparable to baseline designs.