English

Effective Bilevel Optimization via Minimax Reformulation

Machine Learning 2024-11-05 v4 Optimization and Control

Abstract

Bilevel optimization has found successful applications in various machine learning problems, including hyper-parameter optimization, data cleaning, and meta-learning. However, its huge computational cost presents a significant challenge for its utilization in large-scale problems. This challenge arises due to the nested structure of the bilevel formulation, where each hyper-gradient computation necessitates a costly inner optimization procedure. To address this issue, we propose a reformulation of bilevel optimization as a minimax problem, effectively decoupling the outer-inner dependency. Under mild conditions, we show these two problems are equivalent. Furthermore, we introduce a multi-stage gradient descent and ascent (GDA) algorithm to solve the resulting minimax problem with convergence guarantees. Extensive experimental results demonstrate that our method outperforms state-of-the-art bilevel methods while significantly reducing the computational cost.

Keywords

Cite

@article{arxiv.2305.13153,
  title  = {Effective Bilevel Optimization via Minimax Reformulation},
  author = {Xiaoyu Wang and Rui Pan and Renjie Pi and Jipeng Zhang},
  journal= {arXiv preprint arXiv:2305.13153},
  year   = {2024}
}

Comments

Additional experiments and theory update

R2 v1 2026-06-28T10:41:36.551Z