About this Research Topic
Advances in laboratory instrumentation and measurement techniques are helping to provide a wealth of multi-dimensional and multi-modal data to probe materials.
The size and complexity of that data creates challenges in several aspects throughout the whole life cycle of the experimental process. Experiment planning, fast analysis of the data during collection, and final analysis are all made more challenging.
Machine learning has had increased use in commercial applications over the past decade and can offer solutions for experimental physics and materials science. As access to compute resources is made easier, and as machine learning packages are made available to lower the barrier to entry, machine learning is becoming an emerging topic to support physics and materials science research. This Research Topic aims at exploring the applicability of machine learning to help accelerate the experimental process and drive scientific discovery in new directions.
This Research Topic aims to explore the application of machine learning to the modeling of physics and materials science data as well as the planning and execution of experiments.
The use of machine learning has the potential to minimize the time needed to model data using traditional methods and to allow us to solve more complex problems that would otherwise be difficult. Beyond off-line modeling of data, on-the-fly feature extraction can help optimize data collection to maximize scientific output. This can only be made possible by developing a well-integrated experimental environment and closing the loop between instrumentation and analysis. For this purpose, machine learning has the potential to help scientists with experiment design and help facilities orchestrate automated experiments.
The themes for contributions to this Research Topic should cover all aspects needed to make the vision of automated experiments possible. We invite manuscripts addressing, but not limited to, the following:
- Application of machine learning to extract information from complex data sets
- Fault detection and approaches to validate data during acquisition
- Approaches to supplement traditional data analysis with machine learning
- Novel physics-informed methods using materials simulation
- Analyses and workflows to enable autonomous control of experiments
We invite authors to submit manuscripts describing both Original Research and Perspectives.
This Research Topic has been realized in collaboration with Dr. Andrew McCluskey, Instrument Data Scientist at the European Spallation Source.
Keywords: machine learning, materials science, nanomaterials, neutron scattering, x-ray scattering, microscopy
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.