Low-Bitwidth Training

[TCAD'22] Exploring the Potential of Low-bit Training of Convolutional Neural Networks
We propose a hardware-friendly multi-level scaling format to enable low-bit training, and implements the corresponding hardware unit.
Kai Zhong, Xuefei Ning, Guohao Dai, Zhenhua Zhu, Tianchen Zhao, Shulin Zeng, Yu Wang, Huazhong Yang
Last updated on May 7, 2024
PDF Cite DOI
[TCAD'22] Exploring the Potential of Low-bit Training of Convolutional Neural Networks