用位移操作加速边缘设备上的幂次量化，提升性能并降低能耗。
Accelerating PoT Quantization on Edge Devices
- 设计基于位移的处理单元，适配多种幂次量化方法。
- 相比乘法器加速器，平均提速1.23倍、能效提升1.24倍。
- 开源全流程工具链，适合资源受限边缘设备部署。
非均匀量化如幂次（PoT）量化比均匀量化更贴合数据分布，可降低深度神经网络（DNN）的量化误差。PoT量化支持用位移操作替代乘法，但针对位移加速器的效率研究较少。此外，现有在边缘设备上加速PoT量化DNN的流程缺乏开源实现。本文首先为不同PoT量化方法设计位移型处理单元（shift-PE），并通过合成基准测试其效率；随后构建基于最高效shift-PE的位移型加速器，并提出PoTAcc——一个面向资源受限边缘设备的端到端加速开源管道。利用PoTAcc，我们在三个DNN上评估了该加速器性能：平均相较乘法器加速器实现1.23倍速度提升与1.24倍能效改善，较纯CPU执行达2.46倍提速和1.83倍节能。代码已公开于https://github.com/gicLAB/PoTAcc。
原文摘要 · Abstract (English)
Non-uniform quantization, such as power-of-two (PoT) quantization, matches data distributions better than uniform quantization, which reduces the quantization error of Deep Neural Networks (DNNs). PoT quantization also allows bit-shift operations to replace multiplications, but there are limited studies on the efficiency of shift-based accelerators for PoT quantization. Furthermore, existing pipelines for accelerating PoT-quantized DNNs on edge devices are not open-source. In this paper, we first design shift-based processing elements (shift-PE) for different PoT quantization methods and evaluate their efficiency using synthetic benchmarks. Then we design a shift-based accelerator using our most efficient shift-PE and propose PoTAcc, an open-source pipeline for end-to-end acceleration of PoT-quantized DNNs on resource-constrained edge devices. Using PoTAcc, we evaluate the performance of our shift-based accelerator across three DNNs. On average, it achieves a 1.23x speedup and 1.24x energy reduction compared to a multiplier-based accelerator, and a 2.46x speedup and 1.83x energy reduction compared to CPU-only execution. Our code is available at https://github.com/gicLAB/PoTAcc
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。