Yulia Rubanova
Papers
1
Total Citations
7
H-Index
1
About
Yulia Rubanova is a researcher at the forefront of machine learning for physical simulation, with a particular focus on learning complex dynamics using graph neural networks (GNNs). Her work addresses the fundamental challenge of simulating rigid-body interactions, where arbitrary shapes collide—a problem long considered notoriously difficult due to complex geometry and strong non-linearities. In her highly cited 2022 paper, "Learning rigid dynamics with face interaction graph networks" (7 citations), Rubanova introduced a novel GNN-based approach that effectively models these intricate collision dynamics, outperforming traditional physics engines in accuracy and generalizability. This contribution is pivotal for advancing robotic manipulation, virtual environments, and computer graphics, where realistic physical interactions are essential. By demonstrating that GNNs can capture the nuances of rigid-body collisions, Rubanova has opened new pathways for data-driven simulation. Her work stands as a key reference for researchers seeking to bridge the gap between geometric complexity and learned physics, making her a notable voice in the growing field of neural simulation.
Research Focus
Key Achievements
Top Papers
- 1Learning rigid dynamics with face interaction graph networks7 citations · 2022