A Comprehensive Review on Universal Machine Learning Interatomic Potentials has been Published

The core technology of Matlantis, the AI model PreFerred Potential (PFP), is one of the Universal Machine Learning Interatomic Potentials (U-MLIP). U-MLIP models learn atomic interactions and are currently being developed by various companies and research institutions for calculating chemical reactions.

A comprehensive review paper on MLIPs has recently been published. This paper aims to provide a practical guide for researchers who are new to U-MLIPs, covering a wide range of topics from the basic structure of MLIPs to their applications in both organic and inorganic systems. Dr. Takamoto from Preferred Networks, a developer of PFP, is also one of the co-authors of this paper.

For more details, please see:
A practical guide to machine learning interatomic potentials – Status and future

A list of our research papers can be found here:
Our Research Papers

Note: There are abbreviations for machine learning potentials such as Neural Network Potential (NNP) and Machine Learning Potential (MLP), but the notation in this text follows the paper’s conventions.