HAQ: Hardware-Aware Automated Quantization With Mixed Precision

Model quantization is a widely used technique to compress and accelerate deep neural network (DNN) inference. Emergent DNN hardware accelerators begin to support mixed precision (1-8 bits) to further improve the computation efficiency, which raises a great challenge to find the optimal bitwidth for...

Mô tả đầy đủ

Chi tiết về thư mục
Những tác giả chính: Wang, Kuan, Liu, Zhijian, Lin, Yujun, Lin, Ji, Han, Song
Tác giả khác: Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Định dạng: Bài viết
Ngôn ngữ:English
Được phát hành: Institute of Electrical and Electronics Engineers (IEEE) 2021
Truy cập trực tuyến:https://hdl.handle.net/1721.1/129522