The majority of the research on the quantization of Deep Neural Networks (DNNs) is focused on reducing the precision of tensors visible by high-level frameworks (e.g., weights, activations, and gradients). However, current hardware still relies on high-accuracy core operations. Most significant is the operation of accumulating products. This high-precision accumulation operation is gradually becoming the main computational bottleneck. This is because, so far, the usage of low-precision accumulators led to a significant degradation in performance. In this work, we present a simple method to train and fine-tune high-end DNNs, to allow, for the first time, utilization of cheaper, 12-bits accumulators, with no significant degradation in accuracy. Lastly, we show that as we decrease the accumulation precision further, using fine-grained gradient approximations can improve the DNN accuracy.
@article{arxiv.2401.14110,
title = {Towards Cheaper Inference in Deep Networks with Lower Bit-Width Accumulators},
author = {Yaniv Blumenfeld and Itay Hubara and Daniel Soudry},
journal= {arXiv preprint arXiv:2401.14110},
year = {2024}
}