Description
Neural networks have been widely used to solve partial differential equations, which also includes solutions of the Schrödinger equation for quantum mechanical problems. The accurate symmetry properties of solutions of the Schrödinger equation are crucial for getting the correct results. However, up to date only permutational symmetry of particles have been considered for electronic wave functions.
Graph neural networks have grown in popularity for modeling chemical systems, due to their ability to conform to the symmetries of molecules. In this project, student will learn and implement graph neural network for vibrational wave functions of small molecules (e.g., water, ammonia), that transform according to irreducible representations of molecular symmetry group and represent solutions of the corresponding vibrational Schrödinger equation.
Special Qualifications:
background in machine learning and programming with Python
| Field | A2: Molecular sciences (application oriented) |
|---|---|
| DESY Place | Hamburg |
| DESY Division | FS |
| DESY Group | CFEL-CMI |
