Knowledge Discovery Lab

The Knowledge Discovery Lab (KDL) studies how to construct causal models of complex systems, a fundamental research challenge at the frontier of machine learning. In particular, we create new methods, algorithms, and systems that infer causal dependence from observational and experimental data about complex and time-varying relationships among people, places, things, and events. Current research focuses on several areas, including: (1) using causal models to provide human-understandable explanations of how deep neural networks make inferences; (2) using causal models to assess the competence of machine learning models (the circumstances under which the models will perform well or poorly); (3) learning causal models that provide accurate inferences when presented with novel inputs; and (4) methods for effective evaluation of methods for causal modeling.