Yesukhei Jagvaral
Papers
1
Total Citations
5
H-Index
1
About
Dr. Yesukhei Jagvaral is a rising star at the intersection of generative modeling and geometric deep learning, with a focus on developing mathematically rigorous frameworks for non-Euclidean data. His most cited work introduces a unified framework for diffusion generative models on the rotation group SO(3), a critical contribution for fields where orientation data is fundamental, such as computer vision and astrophysics. By extending both score-based and denoising diffusion probabilistic models to manifold-valued data, Jagvaral addresses a key limitation of standard Euclidean-based diffusion models, enabling state-of-the-art generation of 3D rotations. This work, already garnering early citations, bridges a significant gap between theoretical machine learning and practical applications in scientific domains. His research pushes the boundaries of generative AI beyond flat geometries, opening new avenues for modeling complex, structured data in the physical sciences. Jagvaral’s contributions are particularly notable for their potential impact on tasks like protein structure prediction, robotics, and cosmological simulations, marking him as a promising young researcher at the forefront of geometric generative modeling.
Research Focus
Key Achievements
Top Papers
- 1