Abstract:Bilevel optimization provides a comprehensive framework that bridges single- and multi-objective optimization, encompassing various formulations, including standard nonlinear programs. This paper focuses on a specific class of bilevel optimization known as simple bilevel optimization. In these problems, the objective is to minimize a composite convex function over the optimal solution set of another composite convex minimization problem. By reformulating the simple bilevel problem as finding the left-most root of a nonlinear equation, we employ a bisection scheme to efficiently obtain a solution that is $\epsilon$-optimal for both the upper- and lower-level objectives. In each iteration, the bisection narrows down an interval by assessing the feasibility of a discriminating criterion. By introducing a novel dual approach and employing the Accelerated Proximal Gradient (APG) method, we demonstrate that each subproblem in the bisection scheme can be solved in ${\mathcal{O}}(\sqrt{(L_{g_1}+2D_z L_{f_1}+1)/\epsilon}|\log\epsilon|^2)$ oracle queries under weak assumptions. Here, $L_{f_1}$ and $L_{g_1}$ represent the Lipschitz constants of the gradients of the upper- and lower-level objectives' smooth components, and $D_z$ is the upper bound of the optimal multiplier of the subproblem. Considering the number of binary searches, the total complexity of our proposed method is ${\mathcal{O}}(\sqrt{(L_{g_1}+2D_z L_{f_1}+1)/\epsilon}|\log\epsilon|^3)$. Our method achieves near-optimal complexity results, comparable to those in unconstrained smooth or composite convex optimization when disregarding the logarithmic terms. Numerical experiments also demonstrate the superior performance of our method compared to the state-of-the-art.

First order necessary optimality conditions for a class of pessimistic bilevel programming problems

Optimality Conditions for Bilevel Programming Problems

First-order Optimality Conditions for Two Classes of Generalized Nonsmooth Semi-Infinite Optimization

New Necessary Optimality Conditions for Bilevel Programs by Combining the MPEC and Value Function Approaches

An Algorithm for the Lower Bounds of Semivectorial Bilevel Programming Problem

Variational analysis for pessimistic semivectorial bilevel programming with nonsmooth data

Functionally Constrained Algorithm Solves Convex Simple Bilevel Problems

Barrier Function for Bilevel Optimization with Coupled Lower-Level Constraints: Formulation, Approximation and Algorithms

Constraint Qualifications for Optimality Conditions and Total Lagrange Dualities in Convex Infinite Programming

Penalty-based Methods for Simple Bilevel Optimization under Hölderian Error Bounds

A Perturbed Value-Function-Based Interior-Point Method for Perturbed Pessimistic Bilevel Problems

Optimality conditions for bilevel programmes via Moreau envelope reformulation*

First-Order Methods for Linearly Constrained Bilevel Optimization

Constraint Qualifications for Convex Inequality Systems with Applications in Constrained Optimization

First-order penalty methods for bilevel optimization

A Conditional Gradient-based Method for Simple Bilevel Optimization with Convex Lower-level Problem

A Near-Optimal Algorithm for Convex Simple Bilevel Optimization under Weak Assumptions

Near-Optimal Convex Simple Bilevel Optimization with a Bisection Method

PLMEs and Disjunctive Decompositions for Bilevel Optimization

Second-Order Optimality Conditions for Mathematical Programs with Equilibrium Constraints