Probabilistic Programming 101
About This Video
Probabilistic programming has become an increasingly popular approach for building models that capture the uncertainty inherent in many real-world applications. By incorporating probabilistic models into software systems, decision-makers can better understand the range of possible outcomes, identify areas of risk and opportunity, and make more informed decisions.
In this hands-on workshop, we will explore the fundamentals of probabilistic programming and demonstrate its value in the context of real-world applications. We will introduce the key concepts and tools needed to build and deploy probabilistic models, including Numpyro, a powerful probabilistic programming framework based on Python.
As a practical example, we will walk through the development of a probabilistic demand model for a retail marketing scenario. We will use Numpyro to implement the model, explore the resulting predictions, and analyze the key drivers of demand.
In This Video
Machine Learning Engineer, Bain & Company, Inc.
Yu Xia is a machine learning engineer with a passion for forecasting, building better models, and solving real-world problems with accuracy and interpretability. She graduated from NYU with a joint major in Mathematics and Data Science, where she honed her skills in statistics, machine learning, and optimization.