Problem
Research questions and friction points this paper is trying to address.
high-dimensional
multi-objective Bayesian optimization
Pareto front
expensive black-box problems
decision variable interactions
High-dimensional Multi-objective Bayesian Optimization with Learned Variable Interactions
🤖 AI Summary
This work addresses the challenge of high-dimensional multi-objective Bayesian optimization (MOBO), where the exponential growth of sampling complexity with dimensionality hinders efficient approximation of the Pareto front. To overcome this limitation, the authors propose ViaMOBO, a novel framework that, for the first time, integrates variable interaction analysis into high-dimensional MOBO. ViaMOBO automatically identifies the separable structure of objective functions without requiring strong assumptions and performs parallel local Bayesian optimization within the resulting subspaces. The method is applicable to both fully and partially separable problems and demonstrates significant performance gains over state-of-the-art approaches on both synthetic and real-world benchmarks, yielding more accurate and sample-efficient approximations of the Pareto front for high-dimensional, expensive multi-objective problems.
📝 Abstract
Multi-objective Bayesian optimization (MOBO) is effective in identifying the Pareto fronts for expensive black-box problems. However, most current MOBO approaches are limited to low-dimensional decision space due to its exponential sampling complexity. This paper presents decision variable interaction analysis-based MOBO, ViaMOBO, a generic framework for expensive multi-objective problems with high-dimensional decision space. The key idea of ViaMOBO is that it utilizes a variable interaction analysis model to determine whether the decision space can be completely or partially divided, and then performs local Bayesian optimization in the divided decision subspaces. Through the variable analysis model, it can be derived whether the objectives in black-box problems are separable, partially separable, or non-separable based on the potential independent or interdependent relationships among decision variables without any strong assumptions. We compare ViaMOBO with the state-of-the-art MOBO methods on both synthetic and real-world benchmarks. The experimental results demonstrate that ViaMOBO outperforms other related MOBO baselines in approximating the Pareto front of high-dimensional expensive multi-objective problems.
Innovation
Methods, ideas, or system contributions that make the work stand out.
high-dimensional optimization
multi-objective Bayesian optimization
variable interaction analysis
Pareto front approximation
separability detection
🔎 Similar Papers
