Rajarshi Roy

Max Planck Institute for Software Systems

Papers

Total Citations

H-Index

About

Rajarshi Roy is a rising researcher at the intersection of artificial intelligence, formal methods, and robotics. His primary research focuses on developing scalable algorithms for learning temporal logic specifications from data—a critical capability for enabling autonomous systems to understand and follow human instructions. Roy’s most-cited work, "Scalable Anytime Algorithms for Learning Fragments of Linear Temporal Logic" (2022, 25 citations), addresses the fundamental challenge of automatically inferring LTL formulas from finite traces. This problem has broad applications in program verification, motion planning, and process mining. His key contribution lies in designing anytime algorithms that can efficiently learn expressive fragments of LTL, making formal specification learning practical for real-world systems. By bridging the gap between symbolic reasoning and data-driven learning, Roy’s work empowers robots and software systems to interpret complex temporal behaviors from examples. His research is particularly impactful for explainable AI and safe autonomy, where learned specifications must be both accurate and human-interpretable. As an early-career researcher, Roy is establishing himself at the forefront of neurosymbolic approaches to formal methods, with his algorithms already influencing work in robotic task planning and automated verification.