Software for Atomistic Machine Learning
The application of machine-learning techniques to atomistic modeling of physics, chemistry and materials science is blooming, and machine learning is becoming an integral part of the toolbox of molecular simulations. As the conceptual framework underlying these techniques become mainstream, the software infrastructure used to apply machine learning to atomistic problems is also evolving, from experimental code to carefully designed, user-friendly, feature-rich and efficient libraries that implement state-of-the-art methods. This special issue welcomes contributions that present a snapshot of this implementation effort, discussing creative solutions of outstanding problems, demonstrating the efficiency and scaling of algorithms, and providing examples of applications to difficult modeling problems.
Software presented in this special issue will need to be “easily available” to academics. There are two aspects of availability that determine how easy it is. The first one is cost: alongside free (as in beer) software, academic versions of software costing a few hundred dollars we also classify as easily available. The second aspect is transparency. The process for obtaining the software needs to be public and not discriminate unduly: the code must be obtainable by all those who are willing to accept simple and conventional licensing terms, without any undue burden of collaboration or constraints on the intended use of the software.
Guest Editors
Gábor Csányi, University of Cambridge
Matthias Rupp, University of Konstanz
Emine Kucukbenli, Boston University and Harvard University
JCP Editors
Michele Ceriotti, EPFL, Institute of Materials
David Manolopoulos, University of Oxford
Angelos Michaelides, University of Cambridge
David Sherrill, Georgia Institute of Technology
More information:
Please note that papers will be published as normal when they are ready in a regular issue of the journal and will populate on a virtual collection page within a few days of publication. Inclusion in the collection will not cause delay in publication.
How to submit:
- Please submit through the online submission system.
- Under manuscript type → select Article or Communication, as appropriate.
- Under manuscript information → Title/Abstract → select “Invited Submission: No”.
- Under manuscript information → Manuscript classification → select Special Topic: “Software for Atomistic Machine Learning”