Causal Bandits Podcast Open Source Causal AI & The Generative Revolution | Emre Kıcıman Ep 16 | CausalBanditsPodcast.com
May 20, 2024
Emre Kıcıman, a core developer of DoWhy, delves into open source causal AI with Microsoft and Amazon collaboration. They discuss the core of science in causal AI, the intersection of language models and world models, the usefulness of modeling physics, and the future of generative AI. The conversation explores challenges in causality, the importance of causal inference in decision-making, and the evolution of libraries promoting causal analysis.
AI Snips
Chapters
Transcript
Episode notes
Correlation vs. Causation
- Emre Kıcıman was struck by researchers presenting strong social science analyses, only to dismiss them due to correlation-causation limitations.
- This inspired him to explore causal inference to address this issue.
Choosing Causal Resources
- Choose causal inference resources based on your background (data science, statistics, etc.).
- Look for resources that prioritize high-level concepts before diving into specific methods.
DoWhy's Four-Step Process
- The DoWhy library's four-step process (model, identify, estimate, refute) makes causal assumptions explicit and promotes robust analysis.
- This structure also simplifies understanding and teaching causal inference concepts.