Human-in-the-Loop Meta Bayesian Optimization for Fusion Energy and Scientific Applications
Abstract
Inertial Confinement Fusion (ICF) holds transformative promise for sustainable, near-limitless clean energy, yet remains constrained by prohibitively high costs and limited experimental opportunities. This paper presents Human-in-the-Loop Meta Bayesian Optimization (HL-MBO), a framework that integrates expert knowledge with few-shot, uncertainty-aware machine learning to accelerate discovery in data-scarce, high-stakes scientific domains. HL-MBO introduces a meta-learned surrogate model with an expert-informed acquisition function to recommend candidate experiments. To foster trust and enable informed decisions, HL-MBO also provides interpretable explanations of its suggestions. We show HL-MBO outperforms current BO methods on ICF energy yield optimization, as well as benchmarks in molecular optimization and critical temperature maximization for superconducting materials.
Related
- Energy-aware frugal Bayesian optimization
- Towards Large-Scale Heterogeneous Data Organization for Scientific Foundation Models: A Nuclear Fusion Case Study
- ADAPT for Science: A Multi-Agent System for Self-Evolving Scientific Optimization
- Transfer learning in bayesian optimization for aircraft design
- A foundation model for energy and radiation systems built on heterogeneous scientific interfaces
BibTeX
@misc{gutierrez2026humanintheloop,
title = {{Human-in-the-Loop Meta Bayesian Optimization for Fusion Energy and Scientific Applications}},
author = {Ricardo Luna Gutierrez and Sahand Ghorbanpour and Rahman Ejaz and Varchas Gopalaswamy and Riccardo Betti and Vineet Gundecha and Aarne Lees and Soumyendu Sarkar},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2605.00068},
eprint = {2605.00068},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.00068},
}
RIS
TY - UNPB
TI - Human-in-the-Loop Meta Bayesian Optimization for Fusion Energy and Scientific Applications
AU - Ricardo Luna Gutierrez
AU - Sahand Ghorbanpour
AU - Rahman Ejaz
AU - Varchas Gopalaswamy
AU - Riccardo Betti
AU - Vineet Gundecha
AU - Aarne Lees
AU - Soumyendu Sarkar
PY - 2026
DA - 2026/09/12
PB - arXiv
DO - 10.48550/arxiv.2605.00068
UR - https://arxiv.org/abs/2605.00068
AB - Inertial Confinement Fusion (ICF) holds transformative promise for sustainable, near-limitless clean energy, yet remains constrained by prohibitively high costs and limited experimental opportunities. This paper presents Human-in-the-Loop Meta Bayesian Optimization (HL-MBO), a framework that integrates expert knowledge with few-shot, uncertainty-aware machine learning to accelerate discovery in data-scarce, high-stakes scientific domains. HL-MBO introduces a meta-learned surrogate model with an expert-informed acquisition function to recommend candidate experiments. To foster trust and enable informed decisions, HL-MBO also provides interpretable explanations of its suggestions. We show HL-MBO outperforms current BO methods on ICF energy yield optimization, as well as benchmarks in molecular optimization and critical temperature maximization for superconducting materials.
ER -