Has anyone any knowledge of machine learning being combined with nuclear fusion data? I ask because I may be pursuing this intersection.
(Anecdotes from HNers can be more insightful than research papers).
(Anecdotes from HNers can be more insightful than research papers).
Quick search found this: Fuzzy logic and support vector machine approaches to regime identification in Joint European Torus (JET), Plasma Science. http://dx.doi.org/10.1109/TPS.2006.875825.
It seems that machine learning could be used to stabilise a nuclear fusion reactor through a real-time control strategy, e.g. optimising the reactor's environment/conditions via micro-adjustments.