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.
@article{arxiv.1910.09876,
title = {Neural Network Training with Approximate Logarithmic Computations},
author = {Arnab Sanyal and Peter A. Beerel and Keith M. Chugg},
journal= {arXiv preprint arXiv:1910.09876},
year = {2026}
}