PyData LA 2018

The talk will explain why data science should embrace an engine for processing cause-effect relationships. I will describe the structure of this engine, how it has revolutionized the data-intensive sciences, and how it is about to revolutions machine learning.

---

www.pydata.org

PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.

PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases.

0:00 Speaker Introduction
1:00 Introduction
1:18 Talk Proverb
2:13 Talk Outline
5:11 Causal Models and the Cognitive Revolution
10:50 Typical Causal Questions & The Limitation of Standard Grammar of Science
18:15 The Ladder of Causation (3 Level Hierarchy)
26:36 Simpson's Paradox
31:00 Explainability Deep-Learning Style
34:26 Distinguish Seeing from Doing
36:19 The Two Fundamental Laws of Causal Inference
38:15 Reading Independencies
40:02 Structural Causal Model (SCM) Inference Engine
42:33 The Seven Pillars of Causal Wisdom
44:44 Pillar 5: External Validity and Sample Selection Bias
45:51 Pillar 5: The Problem in Real Life
47:06 Pillar 5: The Problem in Mathematics
49:16 Conclusion
51:56 Q&A 1: Opinion on Natural Experiments to Discover Causal Connections in Data
57:38 Q&A 2: Opinion on the Popularization of Statistics in News Media
1:01:30 Q&A 3

S/o to https://github.com/trfore for the video timestamps!

Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/numfocus/YouTubeVideoTimestamps

(From Youtube)