Martin Popel, Ondrej Bojar. Training Tips for the Transformer Model. Prague Bull. Math. Linguistics, 110:43-70, 2018. [doi]
@article{PopelB18, title = {Training Tips for the Transformer Model}, author = {Martin Popel and Ondrej Bojar}, year = {2018}, url = {http://ufal.mff.cuni.cz/pbml/110/art-popel-bojar.pdf}, researchr = {https://researchr.org/publication/PopelB18}, cites = {0}, citedby = {0}, journal = {Prague Bull. Math. Linguistics}, volume = {110}, pages = {43-70}, }