Materials AI
The Materials AI category covers the field that applies artificial intelligence to research on understanding and predicting materials at the atomic level. Its theoretical foundation consists of topics that would be covered in solid state physics, quantum chemistry, and related fields. Major research areas of materials AI include machine learning potentials, which approximate the interactions between atoms using neural networks and similar models; inverse design, which searches for the structures and compositions of materials with desired properties; and self-driving laboratories (SDL), which combine artificial intelligence and robotics to automate experiments from design through execution and analysis.
Basics
- 🔒(26/11/05)Atomic Neighborhood
- 🔒(26/11/13)Symmetry of Atomic Systems
- 🔒(26/11/15)Periodic Boundary Conditions
- 🔒(26/11/07)Interatomic Potentials
- 🔒(26/11/11)Potential Energy Surfaces
Solid State Physics
Density Functional Theory
- 🔒(26/11/09)Born–Oppenheimer Approximation
Machine Learning Interatomic Potentials
- 🔒(26/11/17)Machine Learning Interatomic Potentials
Descriptor-Based
- 🔒(26/11/19)Atomic Descriptors
- 🔒(26/11/21)Atom-Centered Symmetry Functions
- 🔒(26/11/25)SOAP Descriptors
- 🔒(26/11/23)Behler–Parrinello Neural Networks (BPNN)
- 🔒(26/11/27)Gaussian Approximation Potentials (GAP)
- 🔒(26/12/01)Spectral Neighbor Analysis Potentials (SNAP)
- Moment Tensor Potential
- 🔒(26/12/03)Deep Potential Molecular Dynamics (DeePMD)
- 🔒(26/11/29)Atomic Cluster Expansion (ACE)
Materials Databases
- The Materials Project (MP): Provides data on the properties of various inorganic materials. (2026.02.02)
Key References
- Steven H. Simon. The Solid State Basics (2019)