Numerical, Mathematical and Machine Learning Models in Science and Technology of Space and Matter

Dear Colleagues,

During the last decades, and thanks to the advancement of computing capabilities, big data as well as machine and deep learning models have demonstrated their utility in different fields of science and technology. Two areas where these methodologies have arrived to stay are astronomy and particle physics.

In the case of astronomy, the increase in data size and complexity has led to this field being in need of data-driven methods that support scientists as an auxiliary tool to the most traditional model-driven approaches. Nowadays, there are successful examples of the application of supervised learning methods, such as support vector machines, decision trees, probabilistic random forest, artificial neural networks, and unsupervised learning methods such as clustering algorithms, dimensionality reduction algorithms, autoencoders, and those for self-organizing maps and outlier detection.

For example, the use of machine learning in astrostatistics has achieved results of real interest. In the case of astronomic observation, the latest developments of adaptive optics for telescopes include the use of deep learning models for the design of tomographic reconstructors, and it seems that they will continue to play a key role in the case of future extra-large telescopes.

In high-energy physics, neural networks and machine learning technologies have already been used for decades due to the need to deal with the huge volumes of data produced in particle colliders and the evolving complexity of detectors. In present times, due to the availability of larger (HPC clusters), faster (GPUs, FPGAs) or more specialized (TPUs) computing resources, the development of more efficient and usable software frameworks and the fast development of new and promising techniques is producing a huge increase in the use of deep learning models, extending to areas beyond data analysis. For example, there are attempts to improve the traditional algorithms for data reconstruction, detector trigger systems, particle and event classification, data quality monitoring, or even MC simulation at generator level.

This Special Issue is justified as, nowadays, there remain numerous open challenges, including deepening the mathematical foundations of all these methodologies. Not only is the development of new models that can cover real needs required but, also, some theorical aspects are still unclear. Topics such as alternative methods to stochastic gradient for the training of neural networks, development of new cost functions able to reduce overfitting, and methodologies for the selection of the best neural networks topology are examples of some of the issues that remain open.

The objective of this Special Issue is to bring together articles on new theoretical advances and their applications to astronomy and particle physics, giving more light to their theorical foundations.

Dr. Fernando Sánchez Lasheras
Prof. Dr. Maria Luisa Sanchez Rodríguez
Prof. Dr. Javier Fernández Menéndez
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• big data
• deep learning
• machine learning
• theory and practice
• astronomy
• astrostatistics
• adaptive optics
• solar observation
• high-energy physics
• event generation
• detector trigger
• data quality validation and monitoring