Meta-Black-Box Optimization Can Do Search Guidance for Expensive Constrained Multi-Objective Optimization
2026-05-11 • Neural and Evolutionary Computing
Neural and Evolutionary Computing
AI summaryⓘ
The authors developed a new method called MetaSG-SAEA to solve tricky optimization problems that have multiple goals and strict rules, which are also costly to evaluate. Their approach guides a low-level algorithm using a meta-policy that focuses on where to search in the solution space, not just how to search. They created a way to simplify constraint evaluations into an ordered scale (called MM-CCI) and used it to help initialize populations for the search. Their experiments show that this method is better than current leading methods and can work well across different types of problems.
Meta-Black-Box OptimizationMulti-Objective OptimizationConstrained OptimizationSurrogate-Assisted Evolutionary AlgorithmMeta-policyConstraint CalibrationPopulation InitializationDiffusion-based InitializationAttention MechanismProblem Generalization
Authors
Yukun Du, Haiyue Yu, Jiang Jiang, Shuaiwen Tang, Xiaotong Xie, Haobo Liu, Chongshuang Hu, Shengkun Chang
Abstract
Existing Meta-Black-Box Optimization (MetaBBO) methods focus on how to search when controlling optimizers, but largely overlook where to search. We propose MetaSG-SAEA, a bi-level MetaBBO framework for expensive constrained multi-objective optimization problems (ECMOPs), in which a meta-policy provides search guidance to the low-level Surrogate-Assisted Evolutionary Algorithm (SAEA). To achieve this, we introduce Max-Min Constraint-Calibrated Inequality (MM-CCI), a compact, problem-agnostic region abstraction that maps heterogeneous constraint evaluations to an ordered scalar level; we further provide a theoretical analysis of its fundamental properties. Building on this region abstraction, we adopt diffusion-based population initialization to translate the meta-policy's region-level guidance into solution-level priors for the SAEA. To make MetaSG-SAEA scalable, we construct an attention-based state representation across varying problem dimensions, population sizes, and numbers of objectives and constraints. Experimental results demonstrate that MetaSG-SAEA outperforms state-of-the-art baselines across diverse benchmarks and exhibits the ability to generalize across problem distributions.