@INPROCEEDINGS{9053015,

  author={A. {Sanyal} and P. A. {Beerel} and K. M. {Chugg}},
  booktitle={ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, 
  title={{Neural Network Training with Approximate Logarithmic Computations}}, 
  year={2020},
  volume={},
  number={},
  pages={3122-3126},
  abstract={The high computational complexity associated with training deep neural networks limits online and real-time training on edge devices. This paper proposed an end-to-end training and inference scheme that eliminates multiplications by approximate operations in the log-domain which has the potential to significantly reduce implementation complexity. We implement the entire training procedure in the log-domain, with fixed-point data representations. This training procedure is inspired by hardware-friendly approximations of log-domain addition which are based on look-up tables and bit-shifts. We show that our 16-bit log-based training can achieve classification accuracy within approximately 1% of the equivalent floating-point baselines for a number of commonly used datasets.},
  keywords={Logarithmic Number System;Deep Neural Networks;Approximate Computation},
  doi={10.1109/ICASSP40776.2020.9053015},
  ISSN={2379-190X},
  month={May},
  url={https://doi.org/10.1109/ICASSP40776.2020.9053015}
  }