Elias Bareinboim ✓

@eliasbareinboim · New York, NY · joined 22 Jan 2010

Professor of Causal Inference, Machine Learning, and Artificial Intelligence. Director, CausalAI Lab @ Columbia University.

14 847Followers

602Following

2 500Posts total

84.6KViews on collected posts

โพสต์ล่าสุด

Elias Bareinboim ✓@eliasbareinboim7 Jul

Looking forward to reconnecting with many of you and meeting new colleagues at #ICML in Seoul. Across several of our papers this year, we ask a recurring question: What kind of causal knowledge is needed for AI systems to reason beyond the data they directly observe, including 5.2K views · 35 likes · 5 reposts · 0 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim22 Jun

I am back in Israel for the next couple of weeks and would be happy to see friends, colleagues, and students. Please feel free to reach out if you are around, or to share with others who might be interested. I will be giving a few talks on Causal AI and why causality and 5K views · 39 likes · 3 reposts · 3 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim30 Apr

A key point here is that counterfactuals are not only tools for explanation or communication. For a long time, it was natural to think that Layer 2 of the Pearl Causal Hierarchy (PCH) was the layer of decision-making, while Layer 3 was mainly about explanation, attribution, 6.9K views · 47 likes · 9 reposts · 2 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim30 Apr

@FinkelUriah @eliasbareinboim @f2harrell The critical role of counterfactuals in decision making is illustrated in Book of Why p. 361, mainly as a means of communicating knowledge among agents. Moreover, In his insightful book Foresight and Understanding (1961), the philosopher S 13.7K views · 63 likes · 11 reposts · 8 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim20 Mar

Interesting thread. One key point that often gets overlooked: even if models learned continuously post-deployment, interaction alone doesn't resolve the core issue. The same transition data can arise from different causal mechanisms that behave identically under observed 15.4K views · 59 likes · 14 reposts · 5 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim12 May 2025

@eliasbareinboim @sanghack @JunzheZhang12 Thanks again for your visit last week! (posting here as it doesn't seem possible to send you a DM) 292 views · 1 likes · 0 reposts · 1 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim6 Feb 2025

@eliasbareinboim @sanghack @JunzheZhang12 Superb 434 views · 2 likes · 0 reposts · 0 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim5 Feb 2025

@eliasbareinboim @sanghack @JunzheZhang12 Unfortunately, promoting "this Pearlinan epistemiology" in decision making is promoting a pseudoscientific and philosophical vision. This is not improve decision-making, is the opposite in fact. https://t.co/Ti8XYvNUnN

1.4K views · 1 likes · 0 reposts · 1 replies Open on X →

Elias Bareinboim ✓@eliasbareinboim5 Feb 2025

If you are interested in how to improve decision-making, and more fundamentally, how causal inference & RL are related, check out our "Intro to Causal RL" paper: https://t.co/nB3u5CJnDM (with @sanghack & @JunzheZhang12). This follows our ICML tutorial: https://t.co/YbBAXYcyZ7 36.3K views · 264 likes · 35 reposts · 11 replies Open on X →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

36.3K5 Feb

1.4K

4346 Feb

29212 May

15.4K20 Mar

13.7K30 Apr

6.9K

5K22 Jun

5.2K7 Jul

Last 9 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.86%5 Feb

0.22%

0.46%6 Feb

0.68%12 May

0.52%20 Mar

0.62%30 Apr

0.87%

0.91%22 Jun

0.77%7 Jul

Reactions — likes, reposts, replies and quotes — divided by views.

What the audience does

Likes51.0%511 in total

Reposts7.7%77 in total

Replies3.1%31 in total

Quotes1.2%12 in total

Bookmarks37.0%370 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

โพสต์ล่าสุด

Growth & engagement

Views per post

Engagement rate per post

What the audience does

บัญชีที่คล้ายกัน