Kai-Li Chang is a PhD student in the Genetics and Genomics program at Baylor College of Medicine. His research focuses on developing interpretable, efficient, and scalable machine learning models for analyzing single-cell transcriptomics data, with the goal of uncovering data-driven biological mechanisms underlying complex diseases.
He is especially motivated by problems where computational models can generate testable biological hypotheses rather than black-box predictions.
Recently, he developed DiRL (Differentiation with Reinforcement Learning), a novel framework that models cellular differentiation as a sequential decision-making process. DiRL enables in silico perturbation experiments to identify actionable genetic regulators that drive cell state transitions, providing a principled way to explore alternative cell fates and regulatory mechanisms.