-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Atomic simulation is an important theoretical tool for people to understand the spectrum, reaction kinetics, and energy/charge transfer process of complex chemical, biological and material systems at the micro level.
The key element is the need for accurate and efficient high-dimensional potential energy surfaces (force fields)
.
In recent years, the machine learning framework of the Atom Center has become the basic framework for constructing the force field of complex chemical systems
Professor Bin Jiang’s research group has long been committed to the development of high-precision machine learning force field methods
.
Inspired by the combination of atomic orbitals in quantum chemistry into molecular orbitals, the research team based on the embedded atom charge density descriptor proposed earlier, made the orbital coefficients also depend on the surrounding chemical environment, recursively update the orbital coefficients, and developed recursive embedded atoms.
This research work links the neural network based on physical descriptors and the graph neural network based on mathematical models, and provides a new research idea for the development of more accurate and efficient machine learning models
Figure 1 Schematic diagram of recursively embedded atomic neural network model (left) and its comparison with traditional atomic neural network model (right)