-
Biosketch
Hrithik Suresh is a researcher currently pursuing his Ph.D., with a focus on advancing methods in probabilistic modeling and generalization. He was born and brought up in Kottayam district, Kerala. He pursued his undergraduate and master’s studies at the Indian Institute of Science Education and Research (IISER) Thiruvananthapuram, where he gained rigorous training in mathematics, statistics, and machine learning, along with exposure to research-driven problem solving.
His current research interests lie in generalization theory and understanding the role of optimization geometry in machine learning models.
-
Research
My research focuses on probabilistic circuits, investigating how the geometry of the log-likelihood landscape affects generalization. I develop curvature-based measures for PC log-likelihood surfaces to relate sharpness to overfitting. My work integrates theory, algorithm design, and density-estimation experiments on structured PCs. Currently, I develop sharpness-aware training and geometry-based regularization efficiently for probabilistic models.
-
Teaching
- Lead Instructor for a workshop on Python for Data Science, conducted as part of Research Scholars’ Day (RSD) at IIT Palakkad.
- Phd Supervisor
- Research Group
-
Additional Information
TitlePublicationsDescription
Suresh, Hrithik, Sahil Sidheekh, Vishnu Shreeram M P, Sriraam Natarajan, and Narayanan Chatapuram Krishnan. 2026. “Tractable Sharpness-Aware Learning of Probabilistic Circuits”. Proceedings of the AAAI Conference on Artificial Intelligence 40 (30):25736-44. https://doi.org/10.1609/aaai.v40i30.39771.
