About this Research Topic
In the last decade, machine learning has achieved unprecedented results in industrial applications. In particular it is evident that the combination of massive datasets and GPU computing can achieve performances comparable or superior to human capability, in tasks such as image classification and game playing. The scientific community has recently started to look with interest at machine learning and to transfer new methodologies in the workflow of scientific discovery.
In the heliospheric community, machine learning is far from a mainstream approach. However, over the last few years we have seen several pioneering works spanning the whole range of heliophysics: solar physics, solar-terrestrial relationships, magnetospheric, ionospheric and thermospheric physics, and space weather forecasting. This community is now facing the challenge of overcoming the barrier of technical skills not generally mastered by the typical scientist, in order to fully appreciate and critically understand what is within reach in a few years and what could be achieved in a decade.
The first conference on Machine Learning in Heliophysics is taking place in September 2019 in Amsterdam. The scope of the conference is twofold: on one hand, to create a cross-disciplinary research community, formed by physicists in solar, heliospheric, magnetospheric, and aeronomy fields, as well as computer and data scientists. On the other hand, the conference aims at making machine learning more accessible for the whole heliophysics community, thereby lowering the barrier to entry the field.
This Research Topic calls for contributions pertaining to the application of machine learning in any subfield of Heliophysics. Works that have already been presented at the Amsterdam conference are welcome. However, the call is open to all contributors, and not limited to conference participants. We particularly encourage works focused on the process of automation of scientific discovery via machine learning, and on large heliospheric dataset mining. Other relevant topics include:
- inverse estimation of physical parameters,
- automatic event identification,
- feature detection and tracking,
- times series analysis of dynamical systems,
- combination of physics-based models with machine learning techniques,
- surrogate models and uncertainty quantification.
Authors can choose between two type of contributions: 1) a full-length research article (12,000 words and 15 Figures max) or 2) a brief research report (4,000 words and 4 Figures max). In either case the paper is expected to contain novel and original research, and the guest editors will strive to ensure a rapid peer-review and publication timeline.
Keywords: machine learning, data mining, uncertainty qualification, helio19
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.