Mariia Radova

Mariia Radova is a PhD student exploring the application of graph neural networks to understand molecular and materials structure, as well as optimize complex systems like power grids.

PhD Research

Thesis Title: Optimising power grids and chemical reactions with graph neural networks

Supervisors: Prof. Albert Bartok-Partay & Prof. Reinhard Maurer

Mariia’s research addresses a fundamental question: “How do atoms arrange in space to form molecules and materials?” Using graph neural networks (GNNs), she develops computational methods to:

• Predict molecular and materials structures
• Understand chemical transitions between molecules
• Optimize power grid operations
• Design new materials with desired properties

Interdisciplinary Approach

What makes Mariia’s work unique is its interdisciplinary nature, applying the same graph neural network methodologies to both chemical systems and electrical infrastructure. This demonstrates the power of modern machine learning approaches to solve diverse optimization problems across different domains.

Research Impact

Graph neural networks are particularly well-suited for chemistry and materials science because they can naturally represent the connectivity and spatial relationships between atoms. Mariia’s work contributes to the development of next-generation computational tools for:

• Accelerating materials discovery
• Predicting reaction pathways
• Optimizing complex networked systems
• Understanding structure-property relationships

Her research sits at the intersection of chemistry, physics, computer science, and engineering, representing the future of computational science.