Fixed-point feedforward deep neural network design using weights +1, 0, and −1

@article{Hwang2014FixedpointFD,
  title={Fixed-point feedforward deep neural network design using weights +1, 0, and −1},
  author={Kyuyeon Hwang and Wonyong Sung},
  journal={2014 IEEE Workshop on Signal Processing Systems (SiPS)},
  year={2014},
  pages={1-6}
}
Feedforward deep neural networks that employ multiple hidden layers show high performance in many applications, but they demand complex hardware for implementation. The hardware complexity can be much lowered by minimizing the word-length of weights and signals, but direct quantization for fixed-point network design does not yield good results. We optimize… CONTINUE READING

8 Figures & Tables

Topics

Statistics

02040602015201620172018
Citations per Year

119 Citations

Semantic Scholar estimates that this publication has 119 citations based on the available data.

See our FAQ for additional information.