|国家预印本平台
首页|On the Convergence Rates of Iterative Regularization Algorithms for Composite Bi-Level Optimization

On the Convergence Rates of Iterative Regularization Algorithms for Composite Bi-Level Optimization

On the Convergence Rates of Iterative Regularization Algorithms for Composite Bi-Level Optimization

来源:Arxiv_logoArxiv
英文摘要

This paper investigates iterative methods for solving bi-level optimization problems where both inner and outer functions have a composite structure. We provide novel theoretical results, including the first convergence rate analysis for the Iteratively REgularized Proximal Gradient (IRE-PG) method, a variant of Solodov's algorithm. Our results establish simultaneous convergence rates for the inner and outer functions, highlighting the inherent trade-offs between their respective convergence rates. We further extend this analysis to an accelerated version of IRE-PG, proving faster convergence rates under specific settings. Additionally, we propose a new scheme for handling cases where these methods cannot be directly applied to the bi-level problem due to the difficulty of computing the associated proximal operator. This scheme offers surrogate functions to approximate the original problem and a framework to translate convergence rates between the surrogate and original functions. Our results show that the accelerated method's advantage diminishes under this translation.

Shimrit Shtern、Adeolu Taiwo

计算技术、计算机技术

Shimrit Shtern,Adeolu Taiwo.On the Convergence Rates of Iterative Regularization Algorithms for Composite Bi-Level Optimization[EB/OL].(2025-06-19)[2025-07-22].https://arxiv.org/abs/2506.16382.点此复制

评论