.png){kind=icon}
返回
Hardware-Centric AutoML for Mixed-Precision Quantization
.png){kind=icon}
OA
AI
DOI：10.1007/s11263-020-01339-6.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Model quantization is a widely used technique to compress and accelerate deep neural network (DNN) inference. Emergent DNN hardware accelerators begin to supportflexible bitwidth(1-8 bits) to further improve the computation efficiency, which raises a great challenge to find the optimal bitwidth for each layer: it requires domain experts to explore the vast design space trading off accuracy, latency, energy, and model size, which is both time-consuming and usually sub-optimal. There are plenty of specialized hardware accelerators for neural networks, but little research has been done to design specialized neural networks optimized for a particular hardware accelerator. The latter is demanding given the much longer design cycle of silicon than neural nets. Conventional quantization algorithm ignores the different hardware architectures and quantizes all the layers in a uniform way. In this paper, we introduce the Hardware-Aware Automated Quantization (HAQ) framework which automatically determine the quantization policy, and we take the hardware accelerator's feedback in the design loop. Rather than relying on proxy signals such as FLOPs and model size, we employ a hardware simulator to generate the direct feedback signals to the RL agent. Compared with other conventional methods, our framework is fully automated and can specialize the quantization policy for different neural network architectures and hardware architectures. Our framework effectively reduced the latency by 1.4-1.95 x and the energy consumption by 1.9 xwith negligible loss of accuracy compared with the fixed bitwidth (8 bits) quantization. Our framework reveals that the optimal policies on different hardware architectures (i.e., edge and cloud architectures) under different resource constraints (i.e., latency, energy and model size) are drastically different. We interpreted the implication of different quantization policies, which offer insights for both neural network architecture design and hardware architecture design.
Keyword:
Model quantization
Mixed-precision
Automated ML
Hardware
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
9.3
论文数:
3.9K
被引数:
2.8W
机构
暂无机构信息
引用论文
.png){kind=icon}
A Hierarchical Cardiac Rhythm Classification Methodology Based on Electrocardiogram Fiducial Points一种基于心电图基准点的分层心律分类方法OAAI
OAAI分享收藏Do Perceptions of Competence Mediate The Relationship Between Fundamental Motor Skill Proficiency and Physical Activity Levels of Children in Kindergarten?能力的感知是否可以介导幼儿园儿童的基本运动技能熟练程度与身体活动水平之间的关系？
PREAI
分享收藏The power of obfuscation techniques in malicious JavaScript code: A measurement study恶意JavaScript代码中混淆技术的功能: 一项测量研究OAAI
OAAI分享收藏
查看更多内容