Feasibility-Driven Trust Region Bayesian Optimization

2025-06-17

AI Summary

In expensive black-box optimization, the absence of analytical constraint expressions, combined with a small and complex feasible region and difficulty in obtaining initial feasible solutions, often leads to premature budget exhaustion. To address this, we propose a feasibility-driven trust-region Bayesian optimization framework. Our method integrates multi-fidelity surrogate modeling with constraint-aware acquisition, adaptively adjusting the trust region’s center and radius to jointly optimize both objective and constraint surrogates, thereby significantly improving the efficiency of discovering high-quality feasible solutions. Evaluated on the BBOB-constrained COCO benchmark and a physics-inspired test suite across 2–60 dimensions and under varying constraint stringency, our approach consistently outperforms state-of-the-art methods. Notably, it drastically reduces the time to first feasible solution under tight budget constraints, establishing a new, efficient, and robust paradigm for high-cost simulation- and experiment-based optimization.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Optimization
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Bayesian optimization is a powerful tool for solving real-world optimization tasks under tight evaluation budgets, making it well-suited for applications involving costly simulations or experiments. However, many of these tasks are also characterized by the presence of expensive constraints whose analytical formulation is unknown and often defined in high-dimensional spaces where feasible regions are small, irregular, and difficult to identify. In such cases, a substantial portion of the optimization budget may be spent just trying to locate the first feasible solution, limiting the effectiveness of existing methods. In this work, we present a Feasibility-Driven Trust Region Bayesian Optimization (FuRBO) algorithm. FuRBO iteratively defines a trust region from which the next candidate solution is selected, using information from both the objective and constraint surrogate models. Our adaptive strategy allows the trust region to shift and resize significantly between iterations, enabling the optimizer to rapidly refocus its search and consistently accelerate the discovery of feasible and good-quality solutions. We empirically demonstrate the effectiveness of FuRBO through extensive testing on the full BBOB-constrained COCO benchmark suite and other physics-inspired benchmarks, comparing it against state-of-the-art baselines for constrained black-box optimization across varying levels of constraint severity and problem dimensionalities ranging from 2 to 60.

Problem

• Optimizing expensive black-box functions with unknown constraints
• Identifying feasible regions in high-dimensional spaces efficiently
• Accelerating discovery of feasible, high-quality solutions under tight budgets

Innovation

• Feasibility-driven trust region Bayesian optimization
• Adaptive trust region shifting and resizing
• Objective and constraint surrogate model integration