Improving DNN Hardware Accuracy by In-Memory Computing Noise Injection

Sai Kiran Cherupally, Jian Meng, Adnan Siraj Rakin, Shihui Yin, Mingoo Seok, Deliang Fan, Jae-Sun Seo

Abstract

Like any other designs, in-memory computing (IMC) also suffers from operational inaccuracy induced by hardware noise. In this work, the authors propose to take into account hardware noises during the deep neural network (DNN) training in order to improve the DNN inference accuracy.
BibTeX
@inproceedings{cherupally2023improving,
  author = {Sai Kiran Cherupally and Jian Meng and Adnan Siraj Rakin and Shihui Yin and Mingoo Seok and Deliang Fan and Jae-Sun Seo},
  title = {{Improving DNN Hardware Accuracy by In-Memory Computing Noise Injection}},
  booktitle = {IEEE Design \& Test},
  year = {2023},
  url = {https://openreview.net/forum?id=Jh94li2lWD},
}

Loading related content...

Figure