Why Do We Need Large Batchsizes in Contrastive Learning? A Gradient-Bias Perspective

Item #:

068431-2454

Details

DOI: https://doi.org/10.52202/068431-2454
Author(s): Changyou Chen, Liqun Chen, Yiran Chen, Trishul Chilimbi, Jiali Duan, Son Tran, Yi Xu, Belinda Zeng, Jianyi Zhang
Pages: 33860-33875 (16 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 35
Date/Location: Held 28 November - 9 December 2022, New Orleans, Louisiana, USA.
Series: Advances in Neural Information Processing Systems 35
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)

DOI: https://doi.org/10.52202/068431-2454
Author(s): Changyou Chen, Liqun Chen, Yiran Chen, Trishul Chilimbi, Jiali Duan, Son Tran, Yi Xu, Belinda Zeng, Jianyi Zhang
Pages: 33860-33875 (16 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 35
Date/Location: Held 28 November - 9 December 2022, New Orleans, Louisiana, USA.
Series: Advances in Neural Information Processing Systems 35
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)