Bilevel Optimization with Lower-Level Uniform Convexity: Theory and Algorithm

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wu, Yuman, Gong, Xiaochuan, Hao, Jie, Liu, Mingrui
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911474361827328
author Wu, Yuman
Gong, Xiaochuan
Hao, Jie
Liu, Mingrui
author_facet Wu, Yuman
Gong, Xiaochuan
Hao, Jie
Liu, Mingrui
contents Bilevel optimization is a hierarchical framework where an upper-level optimization problem is constrained by a lower-level problem, commonly used in machine learning applications such as hyperparameter optimization. Existing bilevel optimization methods typically assume strong convexity or Polyak-Łojasiewicz (PL) conditions for the lower-level function to establish non-asymptotic convergence to a solution with small hypergradient. However, these assumptions may not hold in practice, and recent work~\citep{chen2024finding} has shown that bilevel optimization is inherently intractable for general convex lower-level functions with the goal of finding small hypergradients. In this paper, we identify a tractable class of bilevel optimization problems that interpolates between lower-level strong convexity and general convexity via \emph{lower-level uniform convexity}. For uniformly convex lower-level functions with exponent $p\geq 2$, we establish a novel implicit differentiation theorem characterizing the hyperobjective's smoothness property. Building on this, we design a new stochastic algorithm, termed UniBiO, with provable convergence guarantees, based on an oracle that provides stochastic gradient and Hessian-vector product information for the bilevel problems. Our algorithm achieves $\widetilde{O}(ε^{-5p+6})$ oracle complexity bound for finding $ε$-stationary points. Notably, our complexity bounds match the optimal rates in terms of the $ε$ dependency for strongly convex lower-level functions ($p=2$), up to logarithmic factors. Our theoretical findings are validated through experiments on synthetic tasks and data hyper-cleaning, demonstrating the effectiveness of our proposed algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2603_00027
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bilevel Optimization with Lower-Level Uniform Convexity: Theory and Algorithm
Wu, Yuman
Gong, Xiaochuan
Hao, Jie
Liu, Mingrui
Optimization and Control
Machine Learning
Bilevel optimization is a hierarchical framework where an upper-level optimization problem is constrained by a lower-level problem, commonly used in machine learning applications such as hyperparameter optimization. Existing bilevel optimization methods typically assume strong convexity or Polyak-Łojasiewicz (PL) conditions for the lower-level function to establish non-asymptotic convergence to a solution with small hypergradient. However, these assumptions may not hold in practice, and recent work~\citep{chen2024finding} has shown that bilevel optimization is inherently intractable for general convex lower-level functions with the goal of finding small hypergradients. In this paper, we identify a tractable class of bilevel optimization problems that interpolates between lower-level strong convexity and general convexity via \emph{lower-level uniform convexity}. For uniformly convex lower-level functions with exponent $p\geq 2$, we establish a novel implicit differentiation theorem characterizing the hyperobjective's smoothness property. Building on this, we design a new stochastic algorithm, termed UniBiO, with provable convergence guarantees, based on an oracle that provides stochastic gradient and Hessian-vector product information for the bilevel problems. Our algorithm achieves $\widetilde{O}(ε^{-5p+6})$ oracle complexity bound for finding $ε$-stationary points. Notably, our complexity bounds match the optimal rates in terms of the $ε$ dependency for strongly convex lower-level functions ($p=2$), up to logarithmic factors. Our theoretical findings are validated through experiments on synthetic tasks and data hyper-cleaning, demonstrating the effectiveness of our proposed algorithm.
title Bilevel Optimization with Lower-Level Uniform Convexity: Theory and Algorithm
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2603.00027