ZeRO: memory optimizations toward training trillion parameter models

Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase, Yuxiong He. ZeRO: memory optimizations toward training trillion parameter models. In Christine Cuicchi, Irene Qualters, William T. Kramer, editors, SC '20: The International Conference for High Performance Computing, Networking, Storage and Analysis, Virtual Event / Atlanta, Georgia, USA, November 9-19, 2020. pages 20, IEEE/ACM, 2020. [doi]

Authors

Samyam Rajbhandari

This author has not been identified. Look up 'Samyam Rajbhandari' in Google

Jeff Rasley

This author has not been identified. Look up 'Jeff Rasley' in Google

Olatunji Ruwase

This author has not been identified. Look up 'Olatunji Ruwase' in Google

Yuxiong He

This author has not been identified. Look up 'Yuxiong He' in Google