Hyperparameter Transfer Enables Consistent Gains of Matrix-Preconditioned Optimizers Across Scales

Item #:

085713-4360

Details

DOI: https://doi.org/10.52202/085713-4360
Author(s): Shikai Qiu, Charlie Chen, Hoang Phan, Qi Lei, Andrew Wilson
Pages: 145075-145119 (45 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 38
Date/Location: Held 2-7 December 2025, San Diego, California, USA and 30 November - 5 December 2025, Mexico City, Mexico.
Series: Advances in Neural Information Processing Systems 38
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)

DOI: https://doi.org/10.52202/085713-4360
Author(s): Shikai Qiu, Charlie Chen, Hoang Phan, Qi Lei, Andrew Wilson
Pages: 145075-145119 (45 pages)
Format: PDF Paper Download
Conference: Advances in Neural Information Processing Systems 38
Date/Location: Held 2-7 December 2025, San Diego, California, USA and 30 November - 5 December 2025, Mexico City, Mexico.
Series: Advances in Neural Information Processing Systems 38
Publisher: Neural Information Processing Systems Foundation, Inc. (NeurIPS)