معرفی
Lucas B Janson is an Associate Professor of Statistics and Affiliate in Computer Science at Harvard University, located at Science Center #710. His primary research focuses on high-dimensional inference and statistical machine learning, with applications in genetics, political science, and climatology. He leads the Harvard Statistical Consulting Service, supervising PhD students advising hundreds of researchers annually.
His teaching includes courses such as Statistical Inference I, Introduction to Reinforcement Learning, and Statistical Machine Learning. He has contributed to numerous software packages, including implementations of knockoff filters, floodgate variable importance, and the distilled conditional randomization test (dCRT).
Key research themes include model-free variable importance, conditional independence testing, and safe reinforcement learning. His work emphasizes methodology with rigorous statistical guarantees, particularly in high-dimensional and complex data settings.
Collaborations span interdisciplinary domains, leveraging statistical tools to address real-world challenges. He actively engages in advancing computational methods for statistical inference and their practical deployment in scientific research.
