%0 Conference Paper %1 conf/icml/RyabininDDB23 %A Ryabinin, Max %A Dettmers, Tim %A Diskin, Michael %A Borzunov, Alexander %B ICML %D 2023 %E Krause, Andreas %E Brunskill, Emma %E Cho, Kyunghyun %E Engelhardt, Barbara %E Sabato, Sivan %E Scarlett, Jonathan %I PMLR %K dblp %P 29416-29440 %T SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient. %U http://dblp.uni-trier.de/db/conf/icml/icml2023.html#RyabininDDB23 %V 202

@inproceedings{conf/icml/RyabininDDB23, added-at = {2023-08-28T00:00:00.000+0200}, author = {Ryabinin, Max and Dettmers, Tim and Diskin, Michael and Borzunov, Alexander}, biburl = {https://www.bibsonomy.org/bibtex/2c85f02d49ef8624c6ae7d6fb3ce30399/dblp}, booktitle = {ICML}, crossref = {conf/icml/2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, ee = {https://proceedings.mlr.press/v202/ryabinin23a.html}, interhash = {67a1f73de3dab8bd0dfe277b57c9dde5}, intrahash = {c85f02d49ef8624c6ae7d6fb3ce30399}, keywords = {dblp}, pages = {29416-29440}, publisher = {PMLR}, series = {Proceedings of Machine Learning Research}, timestamp = {2024-04-10T01:43:32.000+0200}, title = {SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient.}, url = {http://dblp.uni-trier.de/db/conf/icml/icml2023.html#RyabininDDB23}, volume = 202, year = 2023 }