Learning Bayesian Statistics

#141 AI Assisted Causal Inference, with Sam Witty

35 snips
Sep 18, 2025
In this engaging discussion, Sam Whitty, the founder of Sorbus AI and a pioneer in causal probabilistic programming, dives into the intricacies of causal inference. He explores his journey from engineering to developing ChiRho, a language that merges mechanistic and data-driven models. Listeners will learn about counterfactual reasoning, the significance of modular design, and practical applications in science and engineering. Sam emphasizes the need for collaboration in transforming causal questions into actionable insights, while also looking ahead at the future of causal AI.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
ANECDOTE

Engineer's Path Into Causal Modeling

  • Sam started as a mechanical engineer and did energy efficiency consulting where he unknowingly performed causal inference.
  • That experience motivated him to bridge mechanistic engineering models with probabilistic machine learning.
INSIGHT

Why Hybrid Mechanistic+Data Models Matter

  • Mechanistic models extrapolate well because they encode basic physical laws, but they omit complexities best learned from data.
  • Combining mechanistic structure with probabilistic learning preserves extrapolation while improving fit to real-world data.
INSIGHT

What Causal Probabilistic Programming Is

  • Causal probabilistic programming adds an intervention operator to probabilistic programs so you can ask what-if questions about actions.
  • It mechanizes interventions and counterfactual reasoning, turning causal problems into probabilistic inference tasks.
Get the Snipd Podcast app to discover more snips from this episode
Get the app