In a new study published in Proceedings of the National Academy of Sciences, researchers from Los Alamos National Laboratory have proposed incorporating more mathematics of quantum mechanics into the structure of machine learning predictions. Using the specific positions of atoms within a molecule, the machine learning model predicts an effective Hamiltonian matrix, which describes the various possible electronic states along with their associated energies.
Compared to traditional Quantum chemistry simulation machine learningThe approach is based on making predictions at very low computational cost. It enables quantitatively accurate predictions regarding Material properties, allows insight into the nature of chemical bonding between atoms, and can be used to predict other complex phenomena, such as how a system responds to perturbations, such as light-matter interactions. The method also provides significantly improved accuracy over traditional machine learning models, and demonstrates success in portability, that is, the model’s ability to make predictions beyond the data that formed the basis of its training.
Quantum mechanics equations provide a roadmap for predicting the properties of chemicals starting with basic scientific theories. However, these equations can quickly become expensive in terms of computer time and energy when used to predict behavior in large systems. Machine learning offers a promising approach to speed up such large-scale simulations. Using machine learning to predict Chemical properties It holds the potential for significant technological advancement, with applications from clean energy to faster pharmaceutical drug design. This is a very active area of research, but most current approaches use simple and heuristic approaches to design machine learning models.
In their study, the researchers demonstrated that machine learning models can mimic the basic structure of the fundamental laws of nature. It can be very difficult to directly simulate these laws. The machine learning approach enables easy-to-compute and accurate predictions in a wide variety of chemical systems.
The improved machine learning model can quickly and accurately predict a wide range of properties of molecules. These methods score very well in important parameters in computational chemistry And to show how deep learning methods can continue to improve by incorporating more data from experiments. The model can also succeed in challenging tasks such as predicting excited state dynamics – how systems behave with high energy levels. This tool is a breakthrough ability in quantum chemistry. It will allow researchers to better understand the reactivity and excitation states of new molecules.
Guoqing Zhou et al, Deep learning dynamically responsive chemical Hamiltonians with quasi-experimental quantum mechanics, Proceedings of the National Academy of Sciences (2022). DOI: 10.1073/pnas.2120333119
US Department of Energy
the quote: Breakthrough in Machine Learning-enhanced Quantum Chemistry Reported (2022, Sep 13) Retrieved on Sep 13, 2022 from https://phys.org/news/2022-09-breakthrough-machine-learning-enhanced-quantum-chemistry. html
This document is subject to copyright. Notwithstanding any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.