LogNet: Energy-efficient neural networks using logarithmic computation

Edward H. Lee; Daisuke Miyashita; Elaina Chai; Boris Murmann; S. Simon Wong

ICASSP 2017accepted0 citations

LogNet: Energy-efficient neural networks using logarithmic computation

Edward H. Lee, Daisuke Miyashita, Elaina Chai, Boris Murmann, S. Simon Wong

Abstract

We present the concept of logarithmic computation for neural networks. We explore how logarithmic encoding of non-uniformly distributed weights and activations is preferred over linear encoding at resolutions of 4 bits and less. Logarithmic encoding enables networks to 1) achieve higher classification accuracies than fixed-point at low resolutions and 2) eliminate bulky digital multipliers. We demonstrate our ideas in the hardware realization, LogNet, an inference engine using only bitshift-add convolutions and weights distributed across the computing fabric. The opportunities from hardware work in synergy with those from the algorithm domain.

BibTeX

@inproceedings{icassp2017_lognetenergyeffi,
  title = {LogNet: Energy-efficient neural networks using logarithmic computation},
  author = {Edward H. Lee and Daisuke Miyashita and Elaina Chai and Boris Murmann and S. Simon Wong},
  booktitle = {ICASSP 2017},
  year = {2017}
}