Stability and Generalization of Bilevel Programming in Hyperparameter OptimizationDownload PDF

21 May 2021, 20:45 (modified: 28 Oct 2021, 15:17)NeurIPS 2021 PosterReaders: Everyone
Keywords: hyperparameter optimization, generalization, bilevel optimization, stability
TL;DR: A generalization bound of bilevel programming in hyperparameter optimization that explains some mysterious behaviours in practice.
Abstract: The (gradient-based) bilevel programming framework is widely used in hyperparameter optimization and has achieved excellent performance empirically. Previous theoretical work mainly focuses on its optimization properties, while leaving the analysis on generalization largely open. This paper attempts to address the issue by presenting an expectation bound w.r.t. the validation set based on uniform stability. Our results can explain some mysterious behaviours of the bilevel programming in practice, for instance, overfitting to the validation set. We also present an expectation bound for the classical cross-validation algorithm. Our results suggest that gradient-based algorithms can be better than cross-validation under certain conditions in a theoretical perspective. Furthermore, we prove that regularization terms in both the outer and inner levels can relieve the overfitting problem in gradient-based algorithms. In experiments on feature learning and data reweighting for noisy labels, we corroborate our theoretical findings.
Supplementary Material: pdf
Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.
Code: https://github.com/baofff/stability_ho
20 Replies

Loading