US Patent Application 17737733. ORBITAL MIXER MACHINE LEARNING METHOD FOR PREDICTING AN ELECTRONIC STRUCTURE OF AN ATOMIC SYSTEM simplified abstract
Contents
ORBITAL MIXER MACHINE LEARNING METHOD FOR PREDICTING AN ELECTRONIC STRUCTURE OF AN ATOMIC SYSTEM
Organization Name
Inventor(s)
Kirill Shmilovich of Milwaukee WI (US)
Ivan Batalov of Pittsburgh PA (US)
Jeremy Kolter of Pittsburgh PA (US)
Mordechai Kornbluth of Brighton MA (US)
Jonathan Mailoa of Cambridge MA (US)
Devin Willmott of Pittsburgh PA (US)
ORBITAL MIXER MACHINE LEARNING METHOD FOR PREDICTING AN ELECTRONIC STRUCTURE OF AN ATOMIC SYSTEM - A simplified explanation of the abstract
This abstract first appeared for US patent application 17737733 titled 'ORBITAL MIXER MACHINE LEARNING METHOD FOR PREDICTING AN ELECTRONIC STRUCTURE OF AN ATOMIC SYSTEM
Simplified Explanation
The patent application describes a machine learning method for predicting the electronic structure of an atomic system.
- The method involves receiving an atomic identifier and position for atoms in the system, as well as a basis set for forming atomic orbitals.
- Atomic orbitals are formed based on the received information.
- The electronic structure of the atomic system is then predicted using the atomic identifier, atom position, and atomic orbitals.
- The machine learning method is highly accurate and efficient in predicting molecular properties.
- It can directly use basis dependent information to predict the electronic structure of molecules.
- The method utilizes a multi-layer perception (MLP) mixer layers within a simple and scalable architecture.
- It achieves competitive accuracies in predicting Hamiltonian, molecular orbital energy, and coefficient.
Original Abstract Submitted
A machine learning (ML) method for predicting an electronic structure of an atomic system. The method includes receiving an atomic identifier and an atomic position for atoms in the atomic system; receiving a basis set including rules for forming atomic orbitals of the atomic system; forming the atomic orbitals of the atomic system; and predicting an electronic structure of the atomic system based on the atom identifier, the atom position for the atoms in the atomic system, and the atomic orbitals of the atomic system. The ML method is capable of extremely accurate and fast molecular property prediction. The ML can directly purpose basis dependent information to predict molecular electronic structure. The ML method, which may be referred to as an orbital mixer model, uses multi-layer perception (MLP) mixer layers within a simple, intuitive, and scalable architecture to achieve competitive Hamiltonian and molecular orbital energy and coefficient prediction accuracies.
