BMPQ: Bit-Gradient Sensitivity-Driven Mixed-Precision Quantization of DNNs from Scratch-Reference-Cited by-同舟云学术

BMPQ: Bit-Gradient Sensitivity-Driven Mixed-Precision Quantization of DNNs from Scratch

Published:2022-03-14 Issue: Volume: Page:
ISSN:
Container-title:2022 Design, Automation & Test in Europe Conference & Exhibition (DATE)
language:
Short-container-title:

Author:

Kundu Souvik¹,Wang Shikai¹,Sun Qirui¹,Beerel Peter A.¹,Pedram Massoud¹

Affiliation:

1. University of Southern California,Electrical and Computer Engineering,Los Angeles,USA

Funder

NSF

DARPA

Publisher

IEEE

Link

Reference26 articles.

1. HAWQ-V3: Dyadic neural network quantization;yao;Int Conference on Machine Learning,2021

2. Mixed precision quantization of ConvNets via differentiable neural architecture search;wu;ArXiv Preprint,2018

3. Activation density based mixed-precision quantization for energy effi-cient neural networks;vasquez;DATE,2021

4. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding;han;ArXiv Preprint,2015

Cited by 8 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Sensi-Bert: Towards Sensitivity Driven Fine-Tuning for Parameter-Efficient Language Model;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14

4. Distributed Edge Machine Learning Pipeline Scheduling with Reverse Auctions;2023 Eighth International Conference on Fog and Mobile Edge Computing (FMEC);2023-09-18