Reproduce “MF-SCBO : Multi-fidelity Scalable Constrained Bayesian Optimization” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Lucas Palazzolo, Mickaël Binois, Laëtitia Giraldi

MF-SCBO extends scalable constrained Bayesian optimization to high-dimensional problems with black-box constraints and multiple objective fidelities. It combines local trust-region search with Gaussian-process surrogates and cost-aware fidelity selection, while assuming constraints are shared across fidelities. On a 20-variable Ackley problem, its predicted variant reached 0.50 ± 0.21 versus SCBO’s 9.40 ± 1.07 after 4,000 evaluations, a representative result, not a general guarantee.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.