AI-Powered Decision Making Under Uncertainty w/ Allen Downey & Chris Fonnesbeck
In this free live workshop with Allen Downey (Olin College/PyMC Labs) and Chris Fonnesbeck (PyMC Labs/Vanderbilt University Medical Center), learn how to develop Bayesian intuition and build powerful probabilistic models using PyMC—so they actually work in the real world.
Making decisions under uncertainty is hard—especially when your data is limited, your outcomes are rare, or your assumptions are hidden. 😭
Note: This workshop dives into Bayesian probabilistic modeling with PyMC—not generative AI or large-language-model technologies—so you’ll sharpen your statistical decision-making skills under uncertainty.
In this hands-on session, you’ll build the tools to make Bayesian modeling reliable:
🧪 Estimate probabilities with informative priors
🎭 Compare alternatives probabilistically with Bayesian A/B testing
✏️ Share strength across groups using hierarchical models
🚨 Evaluate and anticipate rare events using posterior predictive distributions
We’ll work through a concrete use case, a decision-support system that can:
🗂️ Estimate unknown rates from sparse observations
📊 Pool data across many subgroups
📈 Identify and avoid catastrophic sequences of failures
You’ll see how to evaluate whether it:
🧠 Makes robust predictions under uncertainty
⚙️ Balances generalization and specificity
🗣️ Communicates risks and confidence clearly
And how to build this kind of Bayesian mindset into your modeling workflow—so you can reason clearly, act decisively, and manage uncertainty with confidence.
All running locally or on Colab, using PyMC, ArviZ, and Jupyter Notebooks.