Collection

Artificial Intelligence and High-Performance Computing Algorithms for Environmental Research and Risk Quantification

Submission status
Open
Open for submission from
16 March 2023
Submission deadline
30 June 2024

Artificial Intelligence algorithms have widely been applied for describing processes and creating predictive models in Environmental Research. Despite the fast evolution of mathematical frameworks and algorithmic settings, no golden algorithm has generic applicability to all data-sets and performance metrics. Numerous Artificial Intelligence (AI) algorithms have been outperforming past ones; however, empirical rigor is not following such advancements [1–3]. Researchers argue the accuracy of iterative algorithms and the results obtained for a particular problem are not always reproducible [4,5]. Such criticism on artificial intelligence algorithms, which though efficient, lack a solid theoretical background, is aligned to the scope of this special collection of papers in “Stochastic Environmental Research and Risk Assessment” (SERRA) for generic, efficient, and theoretically sound methods [1,2,6].

Furthermore, in cases when a large volume of data is involved in Environmental Research, the usage of a supercomputer is vital, in order to investigate which algorithm performs better. Accordingly, one drawback of approximation algorithms is tuning the models’ hyper-parameters. For instance, in Artificial Neural Networks, parameters such as learning rate, momentum, number of layers and neurons, vastly affect the accuracy as well as generalization capabilities of a model, and a heuristic search is necessary to obtain robust results. Henceforth, computing power is essential to optimise the applied models as well as their hyperparameters extensively. Scaling up the involved algorithms is also important in exploiting the computing power optimally.

We invite contributions involving Machine Learning and High-Performance Computing Algorithms, for Environmental Science. The novelty should reflect a major advancement in the methodological part of Computational Statistics, Machine Learning, Artificial Intelligence, High-Performance Computing, Predictive modelling, Global optimization, or Data-centric algorithms. The fields of application include Earthquake engineering, Geotechnics, Industrial wastes, Bio-friendly materials, Hydroclimatic Risk, Extreme Rainfalls, Water Resources, Climate change and similar topics as described here.

Keywords — Computational Statistics, Machine Learning, Artificial Intelligence, High-Performance Computing, Predictive modelling, Global optimization, Data-centric algorithms, Earthquake engineering, Geotechnics, Industrial wastes, Bio-friendly materials, Hydroclimatic Risk, Extreme Rainfalls, Water Resources, Climate change

How to Submit Manuscripts should be original and written in English. Submission requires that the manuscript has not been submitted for review or publication elsewhere and that it will not be submitted elsewhere while the review process is underway. All papers go through peer review by at least two experts.

Papers should be submitted online here. Please indicate that your manuscript belongs to the ‘AI and High-Performance Computer Algorithms’ special issue in the ‘Additional Information’ tab during the submission process. Details about the preparation of the manuscript can be obtained from the journal's webpage here. Stochastic Environmental Research and Risk Assessment is a hybrid open access journal.

Authors can opt to make their research Open Access (OA) with Open Choice if they wish, in which case a fee is payable. More information can be found here. Visit Springer Nature’s Open Access funding & support services for information about Transformative Agreements, research funders, and institutions that provide funding for Article Processing Charges.

References

1. N. P. Bakas, A. Langousis, M. Nicolaou, and S. A. Chatzichristofis, “Gradient free stochastic training of anns, with local approximation in partitions,” Stochastic Environmental Research and Risk Assessment, 2023. [Online]. Available: https://link.springer.com/article/10.1007/s00477-023-02407-2

2. N. P. Bakas, V. Plevris, A. Langousis, and S. A. Chatzichristofis, “Itso: A novel inverse transform sampling-based optimization algorithm for stochastic search,” Stochastic Environmental Research and Risk Assessment, vol. 36, no. 1, pp. 67–76, 2022. [Online]. Available: https://link.springer.com/article/10.1007/s00477-021-02025-w

3. D. Sculley, J. Snoek, A. Wiltschko, and A. Rahimi, “Winner’s curse? on pace, progress, and empirical rigor,” in Sixth International Conference on Learning Representations Workshop, 2018. [Online]. Available: https://iclr.cc/Conferences/2018/Schedule?showEvent=406

4. M. Hutson, “Artificial intelligence faces reproducibility crisis,” Science, vol. 359, no. 6377, pp. 725–726, 2018.

5. C. Belthangady and L. A. Royer, “Applications, promises, and pitfalls of deep learning for fluorescence image reconstruction,” Nature methods, p. 1, 2019.

6. N. P. Bakas, “Numerical Solution for the Extrapolation Problem of Analytic Functions,” Research, vol. 2019, pp. 1–10, may 2019. [Online]. Available: https://spj.sciencemag.org/research/2019/3903187/

Editors

  • Nikolaos Bakas Nikolaos Bakas , 

    Nikolaos Bakas

    (Lead Guest Editor) is a Senior Data Scientist at the National Infrastructures for Research and Technology, GRNET and an Assistant professor at The American College of Greece, Deree. His research interests comprise the fundamental mathematical modeling of machine learning algorithms, as well as applications in a variety of thematic areas, such as predictive modeling for tabular datasets and time series, hyperparameter tuning, recommender systems, and large language models. Apart from research, he has been a researcher for various industrial applications of AI. https://github.com/nbakas; nibas@grnet.gr; nbakas@acg.edu

  • Savvas Chatzichristofis Savvas Chatzichristofis , 

    Savvas Chatzichristofis

    Vice-Rector and Professor of AI at Neapolis University Pafos, where he directs the Intelligent Systems Laboratory. His research focuses on the intersection of AI, computer vision, and robotics, and he has been involved in multiple R&D projects funded by European and National organizations. He has received numerous distinctions, grants, scholarships, and awards for his contributions, and has published extensively in prestigious academic journals. More info at https://chatzichristofis.info/; s.chatzichristofis@nup.ac.cy

  • Stergios Emmanouil Stergios Emmanouil , 

    Stergios Emmanouil

    Postdoctoral Research Associate at the Eversource Energy Center of the University of Connecticut (Storrs, United States). His current research focuses on, but is not limited to, the development and application of statistical and stochastic approaches toward the modelling of natural processes and engineering systems, for risk assessment, design, and control. For more information, please visit https://ucwater.engr.uconn.edu/person/stergios-emmanouil/; stergios.emmanouil@uconn.edu

  • Stelios Karozis Stelios Karozis  &

    Stelios Karozis

    Research Associate at Environmental Research Laboratory at NCSR “Demokritos”. His research interests are focus in modelling physical systems using deterministic, stochastic and data-driven techniques. He is also an expert in High Performance Computing (HPC) research and development with experience in the installation and management of HPC infrastructure, as well as in the dissemination of HPC best practices in academia and industry. https://inrastes.demokritos.gr/personnel/karozis-stelios/; skarozis@ipta.demokritos.gr

  • Mihalis A. Nicolaou Mihalis A. Nicolaou

    Mihalis A. Nicolaou

    Currently an Assistant Professor at the Computation-based Science and Technology Center at the Cyprus Institute. His research interests lie in machine learning, computer vision, and signal processing, focusing on the analysis and interpretation of multimodal, high-dimensional data, in a wide range of interdisciplinary applications. He has published over 70 research articles and has received several best paper awards.http://mihalisan.cyi.ac.cy/; m.nicolaou@cyi.ac.cy

Articles (20 in this collection)