Accelerating PoT Quantization on Edge Devices

TL;DR

This paper introduces PoTAcc, an open-source shift-based accelerator and pipeline for efficient PoT quantization of DNNs on edge devices, achieving significant speed and energy improvements over traditional methods.

Contribution

The paper designs shift-based processing elements and an accelerator for PoT quantization, and provides an open-source pipeline for end-to-end DNN acceleration on resource-constrained edge devices.

Findings

Achieves 1.23x speedup over multiplier-based accelerators

Reduces energy consumption by 1.24x compared to multiplier-based accelerators

Outperforms CPU-only execution with 2.46x speedup and 1.83x energy reduction

Abstract

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