SWIS -- Shared Weight bIt Sparsity for Efficient Neural Network Acceleration
This work addresses the need for more efficient neural network acceleration in hardware, offering significant performance gains for deployment on commodity devices, though it is incremental as it builds on existing quantization methods.
The paper tackles the problem of efficient neural network inference by introducing SWIS, a quantization framework that improves accuracy and compression; it achieves up to 54.3% accuracy improvement over weight truncation for MobileNet-v2 at 4 bits and provides up to 6x speedup and 1.9x energy improvement over state-of-the-art bit-serial architectures.
Quantization is spearheading the increase in performance and efficiency of neural network computing systems making headway into commodity hardware. We present SWIS - Shared Weight bIt Sparsity, a quantization framework for efficient neural network inference acceleration delivering improved performance and storage compression through an offline weight decomposition and scheduling algorithm. SWIS can achieve up to 54.3% (19.8%) point accuracy improvement compared to weight truncation when quantizing MobileNet-v2 to 4 (2) bits post-training (with retraining) showing the strength of leveraging shared bit-sparsity in weights. SWIS accelerator gives up to 6x speedup and 1.9x energy improvement overstate of the art bit-serial architectures.