- Published on 08 May 2020
Theory of machine learning, deep learning in particular has been witnessing an implosion lately in deciphering the “black-box approaches”. Optimizing deep neural networks is largely thought to be an empirical process, requiring manual tuning of several parameters. Drawing insights into these parameters gained much attention lately. The special issue aims to focus on gaining theoretical insights in the computation and setting of these parameters and solicits original work reflecting the influence of such theoretical framework on experimental results on standard datasets and architectures. It also aims to garner valuable talking points from optimization studies, another aspect of deep learning architectures and experiments. It is in this spirit that the guest editors wish to bridge metaheuristic optimization methods with deep neural networks and solicit papers that focus on exploring alternatives to gradient descent/ascent types methods. Papers with theoretical insights and proofs are particularly sought after, with or without limited experimental validation. The guest editors would welcome cutting-edge research on aspects of deep learning theory used in the fields of artificial intelligence, statistics and data science, theoretical and numerical optimization.
Habitability outside the solar system is an intriguing topic and center of focused research for at least a decade now. Coupled with this, there are advances in the fields of Artificial Life and Complex Adaptive Systems aiming to understand and synthesize life-like systems. The special issue should bring together material understanding design diversity, complexity, and adaptability of life and their rapid influence in areas of engineering and the Sciences. The guest editors wish to solicit ideas from nature and their generalizations from life and their translations into engineering and science.
Data is at the heart of this. Astronomy is a fascinating case study as it had embraced big data embellished by many sky-surveys. The variety and complexity of the data sets at different wavelengths, cadences etc. imply that modeling, computational intelligence methods and machine learning need to be exploited to understand astronomy. The importance of data driven discovery in Astronomy has given birth to an exciting new field known as astroinformatics. The inter-disciplinary study brings together machine learning theorists, astronomers, mathematicians and computer scientists underpinning the importance of machine learning algorithms and data analytic techniques.
The special issue aims to set a unique ground as an amalgamation of the diverse ideas and techniques while staying true to the baseline. The guest editors expect to discuss new developments in modeling, machine learning, design of complex computer experiments and data analytic techniques which can be used in areas beyond astronomical data analysis. Given the horizontal nature of the issue, they hope to disseminate methods that are area-agnostic but currently of interest to the broad community of science and engineering.
Topics of interest include, but are not limited to:
- Exoplanets (discovery, machine classification etc.)
- Unsupervised, semi-supervised, and supervised representation learning
- Representation learning for reinforcement learning
- Metric learning and kernel learning
- Deep learning in astronomy
- MCMC on big data
- Statistical Machine Learning
- Bayesian Methods in Astronomy
- Meta-heuristic and Evolutionary Clustering methods and applications in Astronomy
- Optimization methods
- Swarm intelligence
- Multi-objective optimization
- Dynamical Systems and Complexity
- Information-Theoretic Methods in Life-like Systems
- Predictive Methods for Complex Adaptive Systems and Life-like Systems
- Gravitational Wave Astronomy
Note to Authors: Proof of concept papers, applied on toy data sets are welcome as long as the theory and models are solid. Papers with applications in random area of Engineering or Science, would be “desk-rejected”. Application in Astronomy is a requirement. The submission and review process follow (strict) single blind protocol.
The contributions are solicited in the following categories:
1. Tutorial/Introductory papers on Foundations, such as Statistics, Machine Learning, Optimization, Numerical Approximation. This is by invitation ONLY.
2. Original Research papers in AstroStatistics and AstroInformatics.
The idea is to provide a primer on the basics and advanced research in the area.
The guest editors invite authors to submit their original research and short reviews on the theme of this special issue. Manuscripts should be prepared following the instructions for authors using the latex template of EPJ ST, which can be downloaded here. Articles should be submitted to the Editorial Office of EPJ ST via the submission system at https://articlestatus.edpsciences.org/is/epjst/home.php by selecting "Modeling, Machine Learning and Astronomy" as special issue