Charting Reality with Stochastic Modeling

3 credits · GSB

Many decision-making systems benefit from stochastic models. These models help algorithms and AI agents make sense of uncertain environments and understand how the impacts of various actions propagate. However, choosing the right model, learning it from data, and evaluating whether it adequately captures reality remain daunting challenges. Emphasizing dynamics arising from physical reality and business problems, we will examine core frameworks for framing, learning, using, and evaluating stochastic models. Examples will range from congestion, AI inference, randomized experiments with causal impact, and physical AI.