Cambricon-Q: A Hybrid Architecture for Efficient Training
Lists
Only you see this
Abstract
Deep neural network (DNN) training is notoriously time-consuming, and quantization is promising to improve the training efficiency with reduced bandwidth/storage requirements and computation costs. However, state-of-the-art quantized algorithms with negligible training accuracy loss, which require on-the-fly statistic-based quantization over a great amount of data (e.g., neurons and weights) and high-precision weight update, cannot be effectively deployed on existing DNN accelerators. To address this problem, we propose the first customized architecture for efficient quantized training with negligible accuracy loss, which is named as Cambricon-Q. Cambricon-Q features a hybrid architecture consisting of an ASIC acceleration core and a near-data-processing (NDP) engine. The acceleration core mainly targets at improving the efficiency of statistic-based quantization with specialized computing units for both statistical analysis (e.g., determining maximum) and data reformating, while the NDP engine avoids transferring the high-precision weights from the off-chip memory to the acceleration core. Experimental results show that on the evaluated benchmarks, Cambricon-Q improves the energy efficiency of DNN training by 6.41× and 1.62×, performance by 4.20× and 1.70× compared to GPU and TPU, respectively, with only ⩽ 0.4% accuracy degradation compared with full precision training.
BibTeX
@inproceedings{zhao2021cambricon,
author = {Yongwei Zhao and Chang Liu and Zidong Du and Qi Guo and Xing Hu and Yimin Zhuang and Zhenxing Zhang and Xinkai Song and Wei Li and Xishan Zhang and Ling Li and Zhiwei Xu and Tianshi Chen},
title = {{Cambricon-Q: A Hybrid Architecture for Efficient Training}},
booktitle = {ISCA},
year = {2021},
doi = {10.1109/isca52012.2021.00061},
}Loading related content...