mlsanityData to 7 Oct 2026

Human-in-the-Loop Meta Bayesian Optimization for Fusion Energy and Scientific Applications

arXiv 2605.00068 ↗First posted 30 Apr 2026cs.LG cs.AI physics.plasm-phcited by 1

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 26 of its 55 references that are in the map. First posted on arXiv 30 Apr 2026.

Builds on

26 of 55 references.

References · 26

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  4. 2018Conditional Neural Processescited by 421
  5. 2021Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Developmentcited by 258
  6. 2022Sample Efficiency Matters: A Benchmark for Practical Molecular Optimizationcited by 151
  7. 2022Transformer Neural Processes: Uncertainty-Aware Meta Learning Via Sequence Modelingcited by 126
  8. 2022Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimizationcited by 96
  9. 2021HPO-B: A Large-Scale Reproducible Benchmark for Black-Box HPO based on OpenMLcited by 75
  10. 2022Recent Advances in Bayesian Optimizationcited by 74
  11. 2019Meta-Learning Acquisition Functions for Transfer Learning in Bayesian Optimizationcited by 70
  12. 2022πBO: Augmenting Acquisition Functions with User Beliefs for Bayesian Optimizationcited by 69
14 more
  1. 2018Dirichlet-based Gaussian Processes for Large-scale Calibrated Classificationcited by 50
  2. 2023End-to-End Meta-Bayesian Optimisation with Transformer Neural Processescited by 31
  3. 2023Looping in the Human Collaborative and Explainable Bayesian Optimizationcited by 27
  4. 2020Incorporating Expert Prior in Bayesian Optimisation via Space Warpingcited by 23
  5. 2021Prior knowledge elicitation: The past, present, and futurecited by 23
  6. 2023A General Framework for User-Guided Bayesian Optimizationcited by 22
  7. 2024Explaining Bayesian Optimization by Shapley Values Facilitates Human-AI Collaborationcited by 12
  8. 2020Learning Inconsistent Preferences with Gaussian Processescited by 11
  9. 2024Explainable Bayesian Optimizationcited by 8
  10. 2020Offline Contextual Bayesian Optimization for Nuclear Fusioncited by 7
  11. 2023HypBO: Accelerating Black-Box Scientific Experiments Using Experts' Hypothesescited by 6
  12. 2020Human Strategic Steering Improves Performance of Interactive Optimizationcited by 5
  13. 2022Transfer learning driven design optimization for inertial confinement fusioncited by 5
  14. 2022Transfer Learning as a Method to Reproduce High-Fidelity NLTE Opacities in Simulationscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Prior or Feedback? What an LLM Uses When Adapting Neural Operatorscited by 0