Physics-informed machine learning and its real-world applications
- Submission status
- Closed
- Submission deadline
Advances in machine learning (ML) and deep learning (DL) are undoubtedly enabling significant breakthroughs in all areas of science and technology. ML/DL models, however, do not necessarily obey the fundamental governing laws of physical systems and often fail to describe and predict scenarios beyond the ones they have been trained on. In addition, training deep neural networks requires a huge amount of quality data, which is not always available for scientific problems. To solve these challenges, a new paradigm that integrates physical principles into ML models is emerging: physics-informed machine learning. Incorporating physics into ML models makes it possible to build physically consistent predictive models which are faster to train, more generalizable, interpretable, and trustworthy.
This Collection aims to gather the latest advances in physics-informed machine learning applications in sciences and engineering. Submissions that provide evidence of scalable, robust, and reliable physics-informed machine learning approaches for large-scale, real-world applications are particularly welcome.
Editors
-
Eleni Chatzi
Swiss Federal Institute of Technology (ETH) Zürich, Switzerland
-
Marta D'Elia
Sandia National Laboratories, USA
-
Jian-Xun Wang
University of Notre Dame, USA
