![]() ![]() Nonetheless, it is the domain knowledge and prior information (e.g. In contrast, experiment specific models have a more relevant, yet more narrow, distribution or training data, and thus struggle when encountering data outside of this distribution. Broader models operate without specific domain knowledge, such as predicted phases of an investigated materials or experimental measurement parameters, and can be directly applied to a broad suite of classification challenges without fine tuning 17, 18. In broader approaches, experimentally relevant domain knowledge can be integrated modularly, e.g. Further refinement of classification results could be achieved by model interpretation 9. Classification models are particularly promising, having been developed for a broad scope (classifying crystal system, space group, point group) 10, 11, 12, 15, and specific challenges integrating experimental information 13, 17, 18, 22. Recent progress has been made in using AI for unsupervised XRD dataset decomposition 6, 7, crystal structure classification 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, and integrating the latter with autonomous experimentation 21. for the identification of composition–structure-relationships, is a challenging cognitive task that requires the ability to recognize patterns under the awareness of several constraints 5. Manual analysis of combinatorial datasets, i.e. Innovations in high-throughput and autonomous experimentation 1, 2, 3, 4 are exceedingly increasing the acquisition rate of data, particularly in the case of XRD. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |