More is Less - Byte-quantized models are faster than bit-quantized models on the edge

Pengfei Zhang, Chenxia Han, Eric Lo 0001. More is Less - Byte-quantized models are faster than bit-quantized models on the edge. In Shusaku Tsumoto, Yukio Ohsawa, Lei Chen 0002, Dirk Van den Poel, Xiaohua Hu 0001, Yoichi Motomura, Takuya Takagi, Lingfei Wu, Ying Xie, Akihiro Abe, Vijay Raghavan 0001, editors, IEEE International Conference on Big Data, Big Data 2022, Osaka, Japan, December 17-20, 2022. pages 5632-5638, IEEE, 2022. [doi]

Abstract

Abstract is missing.