Bilevel optimization (BO) is useful for solving a variety of important machine learning problems including but not limited to hyperparameter optimization, meta-learning, continual learning, and reinforcement learning. Conventional BO methods need to differentiate through the low-level optimization process with implicit differentiation, which requires expensive calculations related to the Hessian matrix. There has been a recent quest for first-order methods for BO, but the methods proposed to date tend to be complicated and impractical for large-scale deep learning applications. In this work, we propose a simple first-order BO algorithm that depends only on first-order gradient information, requires no implicit differentiation, and is practical and efficient for large-scale non-convex functions in deep learning. We provide non-asymptotic convergence analysis of the proposed method to stationary points for non-convex objectives and present empirical results that show its superior practical performance.

PDF

In Conference on Neural Information Processing Systems, 2022, New Orleans, LA, December 2022.

@inproceedings{NeurIPS2022-Liu, title={BOME! Bilevel Optimization Made Easy: A Simple First-Order Approach}, author={Bo Liu and Mao Ye and Stephen Wright and Peter Stone and Qiang Liu}, booktitle={Conference on Neural Information Processing Systems, 2022}, month={December}, address={New Orleans, LA}, url="http://www.cs.utexas.edu/users/ai-lab?NeurIPS2022-Liu", year={2022} }