Hardware-Friendly Logarithmic Quantization with Mixed-Precision for MobileNetV2-Reference-Cited by-同舟云学术

Hardware-Friendly Logarithmic Quantization with Mixed-Precision for MobileNetV2

Published:2022-06-13 Issue: Volume: Page:
ISSN:
Container-title:2022 IEEE 4th International Conference on Artificial Intelligence Circuits and Systems (AICAS)
language:
Short-container-title:

Author:

Choi Dahun¹,Kim Hyun¹

Affiliation:

1. Research Center for Electrical and Information Technology Seoul National University of Science and Technology,Department of Electrical and Information Engineering,Seoul,Korea

Funder

National Research Foundation of Korea(NRF)

Publisher

IEEE

Link

http://xplorestaging.ieee.org/ielx7/9869844/9869845/09869994.pdf?arnumber=9869994

Reference27 articles.

1. Mask R-CNN

2. Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving

3. Implementation of Optimal CNN Accelerators for Mobile Devices: Algorithm, Architecture, and Memory System Co-Design

4. Additive powers-of-two quantization: An efficient non-uniform discretization for neural networks;li;ArXiv Preprint,2019

5. Successive Log Quantization for Cost-Efficient Neural Networks Using Stochastic Computing

Cited by 10 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. SRU-Q: Hardware-friendly Stochastic Rounding Unit-based Gradient Quantization for CNN Training;2024 IEEE 6th International Conference on AI Circuits and Systems (AICAS);2024-04-22

2. PQDE: Comprehensive Progressive Quantization with Discretization Error for Ultra-Low Bitrate MobileNet towards Low-Resolution Imagery;2024 IEEE 6th International Conference on AI Circuits and Systems (AICAS);2024-04-22

3. ACC: Adaptive Compression Framework for Efficient On-device CNN Training;2024 IEEE 6th International Conference on AI Circuits and Systems (AICAS);2024-04-22

4. DPQ: dynamic pseudo-mean mixed-precision quantization for pruned neural network;Machine Learning;2024-01-31

5. Mixed Precision Quantization with Hardware-Friendly Activation Functions for Hybrid ViT Models;2024 International Conference on Electronics, Information, and Communication (ICEIC);2024-01-28