Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks

@article{Carmichael2019PerformanceEfficiencyTO,
  title={Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks},
  author={Zachariah Carmichael and Hamed Fatemi Langroudi and Char Khazanov and Jeffrey Lillie and John L. Gustafson and Dhireesha Kudithipudi},
  journal={ArXiv},
  year={2019},
  volume={abs/1903.10584}
}
Deep neural networks (DNNs) have been demonstrated as effective prognostic models across various domains, e.g. natural language processing, computer vision, and genomics. However, modern-day DNNs demand high compute and memory storage for executing any reasonably complex task. To optimize the inference time and alleviate the power consumption of these networks, DNN accelerators with low-precision representations of data and DNN parameters are being actively studied. An interesting research… CONTINUE READING
7
Twitter Mentions

References

Publications referenced by this paper.