Hands-On Amortized Bayesian Inference

🎙️ Speakers:🎙️ Speakers: Thomas Wiecki, PhD, Luca Fiaschi, PhD, and⏰ Time: 15:00 UTC / 8:00 AM PT / 11:00 AM ET / 5:00 PM Berlin

Traditional MCMC methods like NUTS are the gold standard for parameter estimation, but they hit a wall when dealing with intractable likelihoods, complex simulation models, or real-time production demands where sampling cannot afford to take minutes or hours.

Enter Amortized Bayesian Inference (ABI): a paradigm shift where generative AI architectures (such as Normalizing Flows, Diffusion Models, and Flow Matching) are trained offline to learn complex posterior distributions. Once trained, inference becomes a fast forward pass.

In this hands-on webinar, Stefan Radev (Creator of BayesFlow and Assistant Professor at RPI) joins PyMC Labs to demonstrate how to train neural inference networks and integrate them directly with PyMC workflows.

What you'll take away

• The Fundamentals of ABI: How generative AI turns simulation models into instantaneous probabilistic inference engines.
• Circumvent Intractable Likelihoods: How to train Neural Likelihood Estimators in BayesFlow and plug them into PyMC’s JAX backend for MCMC sampling.
• Train Heavy, Deploy Fast: How to shift computational burden offline to achieve sub-second Bayesian inference in production environments.
• Diagnostics & Trust: How to use Simulation-Based Calibration (SBC) and coverage diagnostics to ensure your neural network isn't "hallucinating" posteriors.
• Live Code Demo: A step-by-step walkthrough building, training, and evaluating a BayesFlow model alongside PyMC.

Who should join

• Data Scientists & ML Engineers who want to deploy Bayesian models into real-time production pipelines.
• PyMC & Stan Users looking to model complex simulator-based data without explicit likelihood functions.
• Researchers & Quantitative Analysts interested in cutting-edge applications of Generative AI for model-based inference.

📜 Outline of Talk / Agenda:

• 5 min: Introduction to PyMC Labs and speakers
• 40 min: Panel discussion
• 15 min: Q&A

-------------------------------------------------
💼 About the speakers:

Dr. Thomas Wiecki ( Founder of PyMC Labs)
Co-author of PyMC, the leading platform for statistical data science. To help businesses solve some of their trickiest data science problems, he assembled a world-class team of Bayesian modelers and founded PyMC Labs - the Bayesian AI consultancy. He did his PhD at Brown University studying cognitive neuroscience.
Connect with Thomas: 👉 Linkedin 🧩 Github

Dr. Luca Fiaschi (PyMC Labs Partner, Gen AI Vertical)
Luca helps organizations unlock the value of data and AI. With 15+ years of experience, he’s led and scaled teams at Mistplay, HelloFresh, Alibaba, and Stitch Fix, driving breakthroughs in personalization, marketing optimization, and causal modeling. He holds a PhD in Computer Science from Heidelberg University.
Connect with Luca👉 LinkedIn🧩 Github

Stefan T. Radev, PhD (Assistant Professor, Creator of BayesFlow)
Stefan is an Assistant Professor at Rensselaer Polytechnic Institute and the principal investigator at BayesOps. He is the creator of BayesFlow, an open-source framework for simulation-based inference, and he's passionate about translating complex probabilistic thinking into usable tools for scientists, developers, and anyone who wants to make better decisions with data.

Connect with Stefan 👉 LinkedIn
------------------
📖 Code of Conduct:
Please note that participants are expected to abide by PyMC's Code of Conduct.

Connecting with PyMC Labs:
🌐 Website: https://www.pymc-labs.com/
👥 LinkedIn: https://www.linkedin.com/company/pymc-labs/
🐦 Twitter: https://twitter.com/pymc_labs
🎥 YouTube: https://www.youtube.com/c/PyMCLabs
🤝 Meetup: https://www.meetup.com/pymc-labs-online-meetup/
🎮 Discord: https://discord.gg/MARSCNemw3