Hierarchical Non-Structured Pruning for Computing-In-Memory Accelerators with Reduced ADC Resolution Requirement

The crossbar architecture, which is comprised of novel nano-devices, enables high-speed and energy-efficient computing-in-memory (CIM) for neural networks. However, the overhead from analog-to-digital converters (ADCs) substantially degrades the energy efficiency of CIM accelerators. In this paper, we introduce a hierarchical non-structured pruning strategy where value-level and bit-level pruning are performed jointly on neural networks to reduce the resolution of ADCs by using the famous alternating direction method of multipliers (ADMM). To verify the effectiveness, we deployed the proposed method to a variety of state-of-the-art convolutional neural networks on two image classification benchmark datasets: CIFAR10, and ImageNet. The results show that our pruning method can reduce the required resolution of ADCs to 2 or 3 bits with only slight accuracy loss (similar to 0.25%), and thus can improve the hardware efficiency by 180%.