Super Data Science: ML & AI Podcast with Jon Krohn

613: Causal Machine Learning

Sep 27, 2022

Dr. Emre Kiciman from Microsoft Research discusses causal machine learning vs correlational, his DoWhy library, key inference steps, future of AGI, leveraging social media for social issues, recommended software tools, hiring criteria for researchers.

01:11:54

Episode guests

Emre Kiciman

AI Summary

AI Chapters

Episode notes

Podcast summary created with Snipd AI

Quick takeaways

Causal ML emphasizes causal assumptions to infer cause and effect relationships in decision-making scenarios.

Emre's DoWhy library aids in modeling assumptions, estimating causal effects, and improving confidence in causal inference outcomes.

Causal inference involves key steps like modeling assumptions, identification, estimation, and validation/refutation to ensure robust analysis outcomes.

Deep dives

Causal Machine Learning and its Impacts

Causal machine learning focuses on incorporating domain knowledge to infer cause and effect relationships, emphasizing the importance of causal assumptions. It aims to guide machine learning models to pay attention to the right patterns through stability in cause and effect relationships, differentiating it from conventional machine learning that looks for correlations only. The application of causal ML in decision-making scenarios is highlighted for robustness against data distribution shifts caused by changing policies.

Intro

3min

Evolution of Microsoft and Introduction to Causal Machine Learning

11min

Exploring Assumptions and Causal Machine Learning with a Data Science Platform Ad

2min

Creation of DUI Library for Causal Inference in Machine Learning

12min

Exploring the Capabilities and Risks of Causal Machine Learning

3min

Importance of Domain Knowledge in Causal Machine Learning

18min

Exploring Various Applications of Causal Machine Learning

4min

Applying Causal Methods in Machine Learning and Hiring Practices at Microsoft Research

16min

Exploring Causal Machine Learning

3min

Dr. Emre Kiciman, Senior Principal Researcher at Microsoft Research joins the podcast to share his world-leading knowledge on causal machine learning.

This episode is brought to you by Datalore (datalore.online/SDS), the collaborative data science platform, and by Zencastr (zen.ai/sds), the easiest way to make high-quality podcasts. Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast for sponsorship information.

In this episode you will learn:
• What is causal machine learning? [5:52]
• Causal machine learning vs correlational machine learning [10:10]
• Emre’s DoWhy open-source library [16:17]
• The four key steps of causal inference [21:24]
• How and why Emre’s key steps of causal inference will impact ML [26:36]
• Emre's thoughts on the future of causal inference and AGI [34:09]
• How Emre leverages social media data to solve social problems [38:36]
• What's next for Emre's research [46:02]
• The software tools Emre highly recommends [55:16]
• What he looks for in the data science researchers he hires [58:45]

Additional materials: www.superdatascience.com/613

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Super Data Science: ML & AI Podcast with Jon Krohn

613: Causal Machine Learning

Episode guests