Robust Regression for Automatic Fusion Plasma Analysis based on Generative Modeling
This work addresses a domain-specific challenge in fusion plasma analysis, offering an incremental improvement for researchers in that field.
The paper tackled the problem of fitting noisy temperature and density spatial profiles in fusion plasma experiments by addressing the lack of knowledge about the data's probability distribution, resulting in a generative modeling-based algorithm that outperforms classical heuristic methods in stability and accuracy.
The first step to realize automatic experimental data analysis for fusion plasma experiments is fitting noisy data of temperature and density spatial profiles, which are obtained routinely. However, it has been difficult to construct algorithms that fit all the data without over- and under-fitting. In this paper, we show that this difficulty originates from the lack of knowledge of the probability distribution that the measurement data follow. We demonstrate the use of a machine learning technique to estimate the data distribution and to construct an optimal generative model. We show that the fitting algorithm based on the generative modeling outperforms classical heuristic methods in terms of the stability as well as the accuracy.