Skip to content

Kumar Kshitij Patelकुमार क्षितिज पटेल

I am a Postdoctoral Associate at the Yale Institute for Foundations of Data Science, where I develop machine learning methods with rigorous guarantees for heterogeneous and sensitive data. My research spans federated learning, optimization, differential privacy, game theory, and data markets, with applications including healthcare. And my recent work also explores private diffusion models and efficient LLM training. During spring 2026, I was a Research Fellow at the Simons Institute for the Theory of Computing, UC Berkeley, as part of the semester program on Federated and Collaborative Learning. Before joining Yale, I received my PhD in Computer Science from the Toyota Technological Institute at Chicago (TTIC) in 2025, advised by Nathan Srebro and Lingxiao Wang. Earlier, I studied computer science at IIT Kanpur, working with Purushottam Kar, and spent a year on academic exchange at EPFL, working with Martin Jaggi.

2026–27 academic job marketSeeking tenure-track faculty positions.

Research directions

Selected papers

All publications ↗

* Equal contribution. The IJCAI author list is alphabetical.

Biography

Kumar Kshitij Patel is a Postdoctoral Associate at the Yale Institute for Foundations of Data Science. His research focuses on the foundations of collaborative and federated learning, particularly learning from heterogeneous and sensitive data in settings such as healthcare. He studies distributed optimization, differential privacy, data valuation, and client incentives, with recent work on private diffusion models and efficient LLM training. During spring 2026, he was a Research Fellow at the Simons Institute for the Theory of Computing, UC Berkeley, in the Federated and Collaborative Learning program. Kshitij received his PhD in 2025 from the Toyota Technological Institute at Chicago, advised by Nathan Srebro and Lingxiao Wang. He completed his undergraduate degree in computer science at IIT Kanpur and spent an academic exchange year at EPFL. His work received an IJCAI 2024 Distinguished Paper Award and a Best Student Paper Honorable Mention at the ICML 2023 Federated Learning Workshop.