Step-size Adaptation Using Exponentiated Gradient Updates

31 January 2022

Papers citing "Step-size Adaptation Using Exponentiated Gradient Updates"

7 / 7 papers shown

Title
Adaptive Coordinate-Wise Step Sizes for Quasi-Newton Methods: A Learning-to-Optimize Approach Wei Lin Qingyu Song Hong Xu 84 1 0 25 Nov 2024
MetaOptimize: A Framework for Optimizing Step Sizes and Other Meta-parameters Arsalan Sharifnassab Saber Salehkaleybar Richard Sutton 19 3 0 04 Feb 2024
An Automatic Learning Rate Schedule Algorithm for Achieving Faster Convergence and Steeper Descent Zhao-quan Song Chiwun Yang 19 9 0 17 Oct 2023
Searching for Optimal Per-Coordinate Step-sizes with Multidimensional Backtracking Frederik Kunstner V. S. Portella Mark W. Schmidt Nick Harvey 8 8 0 05 Jun 2023
Federated Hypergradient Descent A. K. Kan FedML 19 3 0 03 Nov 2022
Learning to Optimize Quasi-Newton Methods Isaac Liao Rumen Dangovski Jakob N. Foerster Marin Soljacic 13 4 0 11 Oct 2022
Large-Scale Differentially Private BERT Rohan Anil Badih Ghazi Vineet Gupta Ravi Kumar Pasin Manurangsi 17 131 0 03 Aug 2021