A 127.8TOPS/W Arbitrarily Quantized 1-to-8b ScalablePrecision Accelerator for General-Purpose Deep Learning

📋 论文概要

该论文提出了一种支持任意量化精度（1-8比特）的可扩展精度加速器，用于通用深度学习推理，能效达127.8 TOPS/W。它解决了不同网络层在稀疏性和精度要求上的差异问题，通过减少存储、逻辑和延迟浪费实现高效处理。

💡 主要创新点

• 提出任意量化精度（1-8比特）的可扩展架构，适配不同网络层的精度需求
• 实现127.8 TOPS/W的高能效，通过最大化利用稀疏性和量化减少计算冗余

🏷 关键词

📄 原文摘要

deep learning accelerators has focused on inference tasks to improve performance by means of maximally utilizing sparsity and quantization. Unlike CNN-only networks, however, recent state-of-the-art (SOTA) models consist of multiple blocks of various layers with different layer-by-layer characteristics in sparsity and required precision. This trend presents challenges in building a general accelerator architecture to maximize the benefits from sparsity and quantization, while supporting efficient processing for various models ranging from traditional CNNs to the new models to come in the future. First, there are multiple considerations that include the bottleneck in data bandwidth, as well as the trade-off between sparsity and required precision. The required

👥 作者与机构

with Reduction of Storage, Logic and Latency Waste

Seunghyun Moon1, Han-Gyeol Mun1, Hyunwoo Son2, Jae-Yoon Sim1 Pohang University of Science and Technology, Pohang, Korea Gyeongsang National University, Jinju, Korea 1 2 Research on