Large-scale Riemannian meta-optimization via subspace adaptation

Peilin Yu, Yuwei Wu, Zhi Gao, Xiaomeng Fan, Yunde Jia

Abstract: Highlights•An efficient Riemannian meta-optimization method for large-scale optimization.•A subspace adaptation scheme for parameter sharing in Riemannian meta-optimization.•Reduce memory consumption of optimizing neural networks by six orders of magnitude.

External IDs:doi:10.1016/j.cviu.2025.104306