FxP-QNet: A Post-Training Quantizer for the Design of Mixed Low-Precision DNNs With Dynamic Fixed-Point Representation

量化(信号处理) 计算机科学 推论 计算机工程 深度学习 人工神经网络 浮点型 深层神经网络 水准点(测量) 计算 人工智能 算法 机器学习 大地测量学 地理
作者
Ahmad Shawahna,Sadiq M. Sait,Aiman H. El‐Maleh,Irfan Ahmad
出处
期刊:IEEE Access [Institute of Electrical and Electronics Engineers]
卷期号:10: 30202-30231 被引量:6
标识
DOI:10.1109/access.2022.3157893
摘要

Deep neural networks (DNNs) have demonstrated their effectiveness in a wide range of computer vision tasks, with the state-of-the-art results obtained through complex and deep structures that require intensive computation and memory. Now-a-days, efficient model inference is crucial for consumer applications on resource-constrained platforms. As a result, there is much interest in the research and development of dedicated deep learning (DL) hardware to improve the throughput and energy efficiency of DNNs. Low-precision representation of DNN data-structures through quantization would bring great benefits to specialized DL hardware. However, the rigorous quantization leads to a severe accuracy drop. As such, quantization opens a large hyper-parameter space at bit-precision levels, the exploration of which is a major challenge. In this paper, we propose a novel framework referred to as the Fixed-Point Quantizer of deep neural Networks (FxP-QNet) that flexibly designs a mixed low-precision DNN for integer-arithmetic-only deployment. Specifically, the FxP-QNet gradually adapts the quantization level for each data-structure of each layer based on the trade-off between the network accuracy and the low-precision requirements. Additionally, it employs post-training self-distillation and network prediction error statistics to optimize the quantization of floating-point values into fixed-point numbers. Examining FxP-QNet on state-of-the-art architectures and the benchmark ImageNet dataset, we empirically demonstrate the effectiveness of FxP-QNet in achieving the accuracy-compression trade-off without the need for training. The results show that FxP-QNet-quantized AlexNet, VGG-16, and ResNet-18 reduce the overall memory requirements of their full-precision counterparts by 7.16x, 10.36x, and 6.44x with less than 0.95%, 0.95%, and 1.99% accuracy drop, respectively.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
377发布了新的文献求助10
刚刚
ay发布了新的文献求助20
3秒前
古德猫宁发布了新的文献求助10
3秒前
典雅的芷蕾完成签到,获得积分10
3秒前
5秒前
hg08发布了新的文献求助10
5秒前
5秒前
搁浅完成签到,获得积分10
5秒前
8秒前
377完成签到,获得积分10
8秒前
Kyt发布了新的文献求助10
9秒前
万能图书馆应助Cl1audia采纳,获得10
9秒前
snutcc完成签到,获得积分10
10秒前
漂亮白枫发布了新的文献求助10
10秒前
古德猫宁完成签到,获得积分10
12秒前
7iy完成签到,获得积分10
14秒前
llllhh完成签到,获得积分10
15秒前
善学以致用应助漂亮白枫采纳,获得10
16秒前
17秒前
悟空完成签到,获得积分10
17秒前
Rita发布了新的文献求助10
17秒前
希望天下0贩的0应助Ran采纳,获得10
18秒前
18秒前
完美世界应助呜呜啦啦采纳,获得10
19秒前
Cl1audia完成签到,获得积分10
20秒前
所所应助Fine采纳,获得10
21秒前
CC发布了新的文献求助10
21秒前
眼睛大的傲菡完成签到,获得积分10
23秒前
23秒前
小二郎应助科研通管家采纳,获得10
23秒前
深情安青应助科研通管家采纳,获得10
23秒前
丘比特应助科研通管家采纳,获得10
23秒前
Jasper应助科研通管家采纳,获得10
23秒前
李爱国应助科研通管家采纳,获得30
23秒前
科目三应助科研通管家采纳,获得10
23秒前
23秒前
深情安青应助科研通管家采纳,获得10
23秒前
充电宝应助科研通管家采纳,获得10
24秒前
24秒前
CipherSage应助juphen2采纳,获得10
25秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
‘Unruly’ Children: Historical Fieldnotes and Learning Morality in a Taiwan Village (New Departures in Anthropology) 400
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
基于可调谐半导体激光吸收光谱技术泄漏气体检测系统的研究 350
Robot-supported joining of reinforcement textiles with one-sided sewing heads 320
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3989797
求助须知:如何正确求助?哪些是违规求助? 3531910
关于积分的说明 11255394
捐赠科研通 3270563
什么是DOI,文献DOI怎么找? 1805008
邀请新用户注册赠送积分活动 882157
科研通“疑难数据库(出版商)”最低求助积分说明 809190