@inproceedings{makingasynchronousstochasticgradient, title = {Making Asynchronous Stochastic Gradient Descent Work for Transformers}, author = {Alham Fikri Aji and Kenneth Heafield}, year = {2019}, booktitle = {WS 2019 11}, eprint = {1906.03496}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/1906.03496v1}, }