Constrained Bayesian Optimisation with Multiple Information Sources
2026-07-01 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address the problem of optimizing a system when you don’t know which options are allowed, especially when valid options are rare. They note that traditional methods struggle because they only use real tests of the system. The authors introduce a new approach that also uses extra, cheaper sources of information to guide the search better. Their method balances cost and useful information and works well even if these extra sources are only loosely connected to the real system. Tests show it finds good and allowed solutions faster than older techniques.
Bayesian OptimisationUnknown ConstraintsMulti-source DataMax-value Entropy SearchSurrogate ModelsDesign Space ExplorationInformation GainEvaluation CostConstrained Optimisation
Authors
Hauke Maathuis, Roeland De Breuker, Saullo Castro, Maike Osborne
Abstract
Bayesian Optimisation (BO) under unknown constraints is particularly challenging when feasible regions are small. In such settings, existing methods that typically rely solely on evaluations of the true objective and constraints struggle to efficiently explore the design space. However, many real-world applications offer auxiliary data sources (e.g. surrogate models or simplified simulations) that can support early exploration. Despite this potential, their integration into constrained BO remains largely unexplored. We propose a general multi-source framework that extends constrained Max-value Entropy Search, capturing inter-source correlation while balancing evaluation cost and information gain. Experiments on both synthetic and physics-based benchmarks show that our method efficiently identifies feasible and optimal solutions, even when auxiliary data are only weakly correlated. The proposed approach consistently outperforms existing methods, particularly in early-stage exploration.