Search Publications

Using multiple instance learning for explainable solar flare prediction

Melchior, M.; Huwyler, C.

In this work we leverage a weakly-labeled dataset of spectral data from NASA's IRIS satellite for the prediction of solar flares using the Multiple Instance Learning (MIL) paradigm. While standard supervised learning models expect a label for every instance, MIL relaxes this and only considers bags of instances to be labeled. This is ideally suite…

Solar activity classification based on Mg II spectra: Towards classification on compressed data

Ivanov, S.; Tsizh, M.; Ullmann, D. +2 more

Although large volumes of solar data are available for investigation and study, the vast majority of these data remain unlabeled and are therefore not amenable to modern supervised machine learning methods. Having a way to accurately and automatically classify spectra into categories related to the degree of solar activity is highly desirable and …