Computer Science 692G - S-Simulation & Casual Modeling

This seminar will explore the relationships between simulation modeling and causal modeling, focusing on the potential of these two research fields to enrich each other. From a simulation point of view, causal models could provide methods for enabling new kinds of tractable probabilistic reasoning about models, for learning models directly from data, and for learning high-level abstractions of a detailed mechanistic model. From a causal-modeling viewpoint, simulation modeling could provide a means to enhance the expressiveness of languages for high-level statistical models with finer-grained mechanistic and generative semantics. Topics will include simulation, causal learning and reasoning, model-based RL, simulation metamodeling, probabilistic programming, process mining, and discovery of causal mechanisms.

Open to Graduate Computer Science students only. SECTION 01=3 CREDITS; SECTION 02=1 CREDIT. A PROJECT IS REQUIRED FOR STUDENTS TAKING FOR 3 CREDITS. STUDENTS NEEDING SPECIAL PERMISSION MUST REQUEST OVERRIDES VIA THE ON-LINE FORM: https://www.cics.umass.edu/overrides.