FusionPapers
図版検索トレンドwiki日本の研究
© 2026 FUSIONPAPERS
About法務情報
トップに戻る

Machine learning and Bayesian inference in nuclear fusion research: an overview

A Pavone, A Merlo, S Kwak, J Svensson2023年Plasma Physics and Controlled FusionIF 2.2出版社

This article reviews applications of Bayesian inference and machine learning (ML) in nuclear fusion research. Current and next-generation nuclear fusion experiments require analysis and modelling efforts that integrate different models consistently and exploit information found across heterogeneous data sources in an efficient manner. Model-based Bayesian inference provides a framework well suited for the interpretation of observed data given physics and probabilistic assumptions, also for very complex systems, thanks to its rigorous and straightforward treatment of uncertainties and modelling hypothesis. On the other hand, ML, in particular neural networks and deep learning models, are based on black-box statistical models and allow the handling of large volumes of data and computation very efficiently. For this reason, approaches which make use of ML and Bayesian inference separately and also in conjunction are of particular interest for today's experiments and are the main topic of this review. This article also presents an approach where physics-based Bayesian inference and black-box ML play along, mitigating each other's drawbacks: the former is made more efficient, the latter more interpretable.

日本語訳

抄録: 本稿は、核融合研究におけるベイズ推論と機械学習(ML)の応用を概説するものである。現在および次世代の核融合実験には、異なるモデルを一貫して統合し、多様なデータソースにわたる情報を効率的に活用する解析およびモデリングのアプローチが必要とされる。モデルベースのベイズ推論は、物理的仮定と確率的仮定に基づいて観測データを解釈するための枠組みを提供し、非常に複雑なシステムに対しても、不確実性とモデリング仮定を厳密かつ系統的に扱うことができる。一方、ML、特にニューラルネットワークおよびディープラーニングは、ブラックボックス型の統計モデルに基づき、大規模なデータと計算を効率的に処理することを可能にする。このため、ベイズ推論とMLを個別に、あるいは組み合わせて活用するアプローチは、現在の実験において特に重要であり、本概説の主要なテーマである。さらに本稿では、物理に基づくベイズ推論とブラックボックス型MLが相互に補完し合い、それぞれの欠点を緩和するアプローチも提示する:前者はより効率的になり、後者はより解釈可能になる。

この論文にはまだAI要約がありません。

関連論文

Neural network approximation of Bayesian models for the inference of ion and electron temperature profiles at W7-X

2019Plasma Physics and Controlled Fusion

Decision support for engineering and design in a fusion pilot-plant concept using Bayesian networks as meta-models

2025Nuclear Fusion

Neural network approximated Bayesian inference of edge electron density profiles at JET

2020Plasma Physics and Controlled Fusion

Bayesian approach to parameter estimation and model validation for nuclear fusion reactor mean-field edge turbulence modelling

2021Nuclear Fusion

Integrated data analysis of fusion diagnostics by means of the Bayesian probability theory

2004Review of Scientific Instruments

Integrated data analysis on the electron density profile of HL-2A with the Bayesian probability inference method

2023Plasma Physics and Controlled Fusion

Normalizing flows for likelihood-free inference with fusion simulations

2022Plasma Physics and Controlled Fusion

Development of a Bayesian method for the analysis of inertial confinement fusion experiments on the NIF

2013Nuclear Fusion

Using statistical modeling to predict and understand fusion experiments

2021Physics of Plasmas

Bayesian neural networks for predicting tokamak energy confinement time with uncertainty quantification

2025Nuclear Fusion