Modern Deep Neural Networks (DNNs) require significant memory to store weight, activations, and other intermediate tensors during training. Hence, many models do not fit one GPU device or can be trained using only a small per-GPU batch size. This survey provides a systematic overview of the approaches that enable more efficient DNNs training. We analyze techniques that save memory and make good use of computation and communication resources on architectures with a single or several GPUs. We summarize the main categories of strategies and compare strategies within and across categories. Along with approaches proposed in the literature, we discuss available implementations.
@article{arxiv.2202.10435,
title = {Survey on Large Scale Neural Network Training},
author = {Julia Gusak and Daria Cherniuk and Alena Shilova and Alexander Katrutsa and Daniel Bershatsky and Xunyi Zhao and Lionel Eyraud-Dubois and Oleg Shlyazhko and Denis Dimitrov and Ivan Oseledets and Olivier Beaumont},
journal= {arXiv preprint arXiv:2202.10435},
year = {2022}
}