Meta-Black-Box Optimization with Bi-Space Landscape Analysis and Dual-Control Mechanism for SAEA

2025-11-19

AI Summary

Surrogate-assisted evolutionary algorithms (SAEAs) for multi-objective expensive black-box optimization suffer from poor generalizability due to heavy reliance on handcrafted designs. Method: This paper proposes DB-SAEA, a meta black-box optimization framework that introduces a novel meta-strategy for dual-space landscape analysis—jointly characterizing the surrogate model space and the true objective space—augmented by an attention mechanism to encode high-dimensional multi-objective optimization states. It features a dual-control mechanism jointly optimizing candidate solution generation policies and infill sampling criteria, and integrates TabPFN surrogates within a reinforcement learning framework to enable parallel sampling and centralized meta-training. Contribution/Results: DB-SAEA significantly outperforms state-of-the-art methods on standard benchmarks and demonstrates strong zero-shot transfer capability—directly deploying to unseen high-dimensional tasks without fine-tuning—thereby validating its adaptability, universality, and scalability.

Technology Category

• Search and Optimization: Metareasoning and Metaheuristics
• Intelligent Robots: Learning & Optimization for ROB
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Surrogate-Assisted Evolutionary Algorithms (SAEAs) are widely used for expensive Black-Box Optimization. However, their reliance on rigid, manually designed components such as infill criteria and evolutionary strategies during the search process limits their flexibility across tasks. To address these limitations, we propose Dual-Control Bi-Space Surrogate-Assisted Evolutionary Algorithm (DB-SAEA), a Meta-Black-Box Optimization (MetaBBO) framework tailored for multi-objective problems. DB-SAEA learns a meta-policy that jointly regulates candidate generation and infill criterion selection, enabling dual control. The bi-space Exploratory Landscape Analysis (ELA) module in DB-SAEA adopts an attention-based architecture to capture optimization states from both true and surrogate evaluation spaces, while ensuring scalability across problem dimensions, population sizes, and objectives. Additionally, we integrate TabPFN as the surrogate model for accurate and efficient prediction with uncertainty estimation. The framework is trained via reinforcement learning, leveraging parallel sampling and centralized training to enhance efficiency and transferability across tasks. Experimental results demonstrate that DB-SAEA not only outperforms state-of-the-art baselines across diverse benchmarks, but also exhibits strong zero-shot transfer to unseen tasks with higher-dimensional settings. This work introduces the first MetaBBO framework with dual-level control over SAEAs and a bi-space ELA that captures surrogate model information.

Problem

• Overcoming rigid manual design limitations in surrogate-assisted evolutionary algorithms
• Enhancing flexibility and transferability across multi-objective optimization tasks
• Addressing scalability challenges across problem dimensions and population sizes

Innovation

• Dual-control mechanism jointly regulates generation and selection
• Bi-space landscape analysis captures states from dual spaces
• Reinforcement learning trains meta-policy for cross-task transferability