Efficient Hessian-Free Methods for Multi-Objective Bilevel Optimization with Nonconvex Lower Level

2026-08-12

AI Summary

This work addresses the challenge posed by nonconvex lower-level problems in multi-objective bilevel optimization, which renders existing methods ineffective. To overcome this limitation, the authors propose a single-level equivalent reformulation based on the Moreau envelope and integrate it with a smoothed weighted Tchebycheff scalarization, yielding the first single-loop, Hessian-free solution framework capable of handling nonconvex lower-level subproblems. Building upon this foundation, they further develop a momentum-based stochastic algorithm, MB-MOMEHA, which synergistically combines momentum acceleration with stochastic optimization. Experimental results demonstrate that the proposed method efficiently generates high-quality Pareto fronts in few-shot meta-learning and neural architecture search tasks, significantly outperforming state-of-the-art approaches and confirming its effectiveness and robustness.

Abstract

Multi-objective bilevel optimization has wide applications in the AI area such as automated learning and multi-task meta-learning. Although recently some works have been begun to study the multi-objective bilevel optimization, the proposed methods rely on the (strongly) convex lower level problems. In fact, these multi-objective bilevel learning problems are generally nonconvex, and particularly their lower level problems are nonconvex. To fill this gap, we propose a class of Multi-Objective Moreau Envelope based Hessian-free Algorithms (MOMEHA) to solve the multi-objective bilevel learning problems with nonconvex lower level. Specifically, our method uses the Moreau envelope to convert the original problem into a multi-objective single-level optimization with an envelope constraint. In particular, our method retains computational advantages of being single-loop and Hessian-free in the multi-objective setting by incorporating a smooth weighted Tchebycheff scalarization. Furthermore, we propose a momentum-based variant of MOMEHA (i.e., MB-MOMEHA) method to solve the stochastic multi-objective bilevel learning problems. In theory, we provide the convergence properties of our algorithms under both deterministic and stochastic setting. Some experiments on few-shot meta-learning and neural architecture search demonstrate that our methods outperform the existing approaches in Pareto front, validating its effectiveness and robustness.

Problem

• multi-objective bilevel optimization
• nonconvex lower level
• Hessian-free methods
• meta-learning
• neural architecture search

Innovation

• Hessian-free
• multi-objective bilevel optimization
• Moreau envelope
• nonconvex lower level
• Tchebycheff scalarization