Kasun Guruge

Kasun is a Ph.D. student in Computer Science at New Mexico State University with a research focus on offline reinforcement learning (Offline RL), particularly on developing theoretical foundations and algorithms for reliable policy learning in settings with limited, noisy, or suboptimal data. His goal is to advance the understanding of how agents can learn and generalize effectively without online interactions, addressing challenges such as distributional shift, uncertainty estimation, and constrained decision-making.

His academic journey began with a B.Sc. in Software Engineering, followed by an M.Sc. in Big Data Analytics, during which he explored natural language processing and machine learning applications. Prior to starting my Ph.D., he gained extensive industry experience as a backend and DevOps engineer, working on large-scale systems at companies like Virtusa and Arimac Lanka, and as a machine learning engineer in the UAV R&D wing of the Sri Lanka Air Force, where he developed autonomous navigation and object detection systems.

This combination of theoretical research and practical engineering experience allows him to approach reinforcement learning challenges from both a mathematical and systems perspective. His long-term objective is to contribute to the development of principled Offline RL methods that can be applied across diverse domains, bridging the gap between theory and real-world deployment.