Elias Bareinboim @eliasbareinboim Twitter profile | Pikagi

Pikagi

Elias Bareinboim

@eliasbareinboim

14,081

Followers

560

Following

120

Media

2,259

Statuses

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

New York, NY

https://t.co/RBM4w7eGnW

Joined January 2010

Don't wanna be here? Send us removal request.

Pinned Tweet

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

If you are curious about when neural nets can perform causal inferences, & more fundamentally, how neural & causal models are related, check out: “The Causal-Neural Connection: Expressiveness, Learnability, and Inference”: (with K Xia, K Lee, Y Bengio)

Tweet card media

The Causal-Neural Connection: Expressiveness, Learnability, and Inference

One of the central elements of any causal inference is an object called structural causal model (SCM), which represents a collection of mechanisms and exogenous sources of random variation of the...

8

121

554

Last Seen Profiles

@ushirogata_bkk

@_Junimill

@Futasister_slut

@MissinRoyko

@tqi105

@z1pzlyo34

@marymorrissey

@shin_kei30ASIA

@rikuu2205

@penostalgic

@krng_s3

@GusCatCrypto

@coach_galvan

@amoy3rd

@NotYourFool10

@ruu_burlesque

@FullMetalTools

@PuckPodcast

@bshanawani

@Bena_King_Green

@luxmizumi

@Coysfr

@Kymoniii

@emmanpv

@DarkHorseDirect

@PeterJetten

@Mark81717454

@ITNArchive

@gamja561702

@okrfinancial

@Hail20207

@RinT___T

@Hmv_Hell

@SpiceAndLewds

@martininstitute

@tiagosada

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Interested in Causal Inference & Reinforcement Learning? Consider attending my @icmlconf tutorial on the basic principles & tools of Causal Reinforcement Learning (CRL). I’ll discuss many new & pervasive learning challenges/opportunities within CRL. Link:

6

113

615

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

As promised, here is the video of my talk last week @MIT on causal inference, generalizability, & fusion: . The lecture includes a summary of recent advances in the field. Thanks, @VC31415 et al. for the intellectually stimulating discussions. @mitidss

Tweet card media

IDSS Distinguished Speaker Seminar Series with Elias Bareinboim...

www.youtube.com

2

81

374

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

The WHY-21 workshop "Causal Inference & Machine Learning: Why now?" is currently accepting submissions, . Our goal is to bring CI & ML researchers together to discuss the nextgen AI! (joint w/ @yudapearl , Y. Bengio, T. Sejnowski, @bschoelkopf ) @NeurIPSConf

1

93

351

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

The WHY'21 workshop "Causal Inference & Machine Learning: Why now?" will take place this Monday at #NeurIPS2021 . Our goal is to bring CI & ML researchers together to discuss the nextgen AI! Program (joint w/ @yudapearl , @bschoelkopf , Y Bengio, T Sejnowski)

5

58

326

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

1/5 If you are interested in causal inference & machine learning, I am excited to share some of the latest work of the CausalAI Lab that will appear @ICMLconf this year. We welcome you to stop by and chat during the sessions or at some point during the conference… #ICML2022

2

38

325

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

1/2 Thanks to the 5k new followers in this past year! To celebrate, two announcements -- First, we are beta-testing a tool called ‘Fusion’, which offers an easy-to-use way of doing causal inference from 1st principles (see #bookofWHY ). Subscribe here: .

Tweet card media

Causal Fusion Beta Testing

docs.google.com

6

70

315

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

1/6 If you are interested in causal inference & machine learning, I am excited to share some of the latest work of the causal artificial intelligence group that will appear @NeurIPSConf this year. We welcome you to stop by and chat with us at the following times… #NeurIPS2021

2

51

307

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

1/n Dear friends, I am pleased to announce that after four productive years here at Purdue, I decided to move to Columbia University, starting on this July 1st. I am excited about the new possibilities and adventures that we’ll have in causality-land in the next years, both in

30

12

297

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

Our work on general identifiability was just selected as UAI-19 Best Paper Award (1 out of 450)! The paper provides the conditions necessary to reason with experimental distributions; e.g., one can combine two dists, do(X1) & do(X2), to answer about the joint effect, do(X1, X2).

@VibhavGogate

Vibhav Gogate

5 years

UAI 2019 best paper award to research on Causal Inference: "General Identifiability with Arbitrary Surrogate Experiments" by Sanghack Lee, Juan D. Correa and Elias Bareinboim ( @eliasbareinboim ). #UAI2019 .

0

10

66

4

25

261

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

Hi Judea, thank you for your kindness and for sharing your wisdom & friendship during this long journey. I feel humbled and blessed to have the opportunity to study & research the foundations of causal inference and AI with you.

@yudapearl

Judea Pearl

2 years

Congratulations to @eliasbareinboim for getting tenure at Columbia University. Simultaneously, congratulations go to Columbia University for a securing its leadership in next-generation AI. This advancement strengthens causal inference research with an important academic 1/2

15

14

292

18

9

226

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

1/3 Just made available the video & slides for my #ICML2020 “Causal Reinforcement Learning” (CRL) tutorial: . CRL combines the strengths of Causal Inference(CI) & RL to solve novel/practical decision-making problems that neither approach can tackle alone!

2

42

214

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 months

@jonathanrlarkin Hey Jonathan, cool to see you interested in causality! :) I would say that the starting point for causal researchers is that we are NOT useful in tasks that reside in layer 1 (L1) of Pearl’s hierarchy, i.e., purely predictive tasks (). I believe the…

9

29

218

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

1/6 If you are interested in causal inference & machine learning, I am excited to share some of the latest work of the causal artificial intelligence group that will appear @NeurIPSConf this year. We welcome you to stop by and chat with us at the following times… #NeurIPS2020

2

42

211

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

The call for papers of our causal inference and machine learning symposium is out -- "Beyond Curve Fitting: Causation, Counterfactuals, and Imagination-based AI," which will happen this spring at Stanford. Details: @yudapearl @bschoelkopf @CsabaSzepesvari

1

62

175

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

Happy to share our tutorial's video on "Causal Fairness Analysis" in the @FAccTConference this year (joint work w/ Drago Plecko & @junzhez ): , and the slides: . Special thanks to the organizers, @tiberiocaetano & @zacharylipton .)

Tweet card media

Tutorial Causal Fairness Analysis (ACM FAccT'21)

Tutorial presented by Professor Elias Bareinboim entitled "Causal Fairness Analysis", which took place at the ACM FAccT conference, Mar/4, 2021. This is join...

www.youtube.com

1

28

155

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

If you are in Boston and curious about the relationship between causal inference & reinforcement learning, which I have been calling "causal reinforcement learning", I'll be talking tomorrow morning at MIT Samberg center, 6th Floor. For more details, see

8

9

154

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

New piece on causal inference & the future of AI at the #MITreview -- "What AI still can’t do" by @BrianBergstein -- including comments by @yudapearl , @bschoelkopf , @rlmcelreath .

Tweet card media

What AI still can’t do

Artificial intelligence won’t be very smart if computers don’t grasp cause and effect. That’s something even humans have trouble with.

www.technologyreview.com

3

58

137

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

The slides of Judea's talk - "The Foundations of Causal Inference, with Reflections on ML and AI" - this week at the #WHY19 is available, see ; more coming soon.

2

35

137

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Just made available the video of my 2020 MSR's Frontiers of Machine Learning talk - "On the Causal Foundations of AI", link: . To understand some of the results (incl. PCH & CHT), see joint chapter w/ J. Correa, D. Ibeling, T. Icard:

Tweet card media

"On the Causal Foundations of AI" -- MSR Frontiers of Machine...

Talk by Professor Elias Bareinboim on "On the Causal Foundations of Artificial Intelligence (Explainability & Decision-Making)" , which appeared at MSR's Fro...

www.youtube.com

0

42

130

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

The WHY-19 will be happening from Mar/25-27 @ Stanford. The theme this year is "Beyond Curve Fitting: Causation, Counterfactuals, Imagination-based AI". We have great speakers, including J. Pearl, Y. Bengio, K. Imai, J. Ioannidis. Don't miss!! #bookofwhy

5

18

119

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

For those interested in my talk about "causal data science" presentation at @MIT last week, here is the link to the slides:

@vumaasha

vumaasha

5 years

@eliasbareinboim @MIT @mitidss @MIT_CSAIL Please do share your presentation. Thanks

0

0

3

3

20

117

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

As promised, link to my keynote on the interactive causal learning conf. last week, including my talk + nice Q & A session: . I summarized the current state of the art in causal AI, and tried demystify & contextualize recent progress & real challenges.

Tweet media one

Tweet media two

Tweet media three

Tweet media four

0

24

114

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

2/2 Due to server constraints, be patient, access will be 'raffle' style. Second, we created a Youtube channel w/ causal inference videos. So far we have @yudapearl 's keynote @ WHY-19 & my lecture on causal data science @Columbia . Watch, subscribe & share:

Tweet card media

We'll be posting videos related to causal inference and related topics, including AI, Machine Learning, and Data Science. For more information about our latest results, see https://causalai.net.

www.youtube.com

1

22

110

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

@tdietterich @yudapearl @roydanroy Thanks, Tom. The video, slides, and other resources are available on the causal RL website:

2

24

110

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

1/6 If you are curious about how causality & imitation learning are related, check our latest #NeurIPS2020 paper (w/ @junzhez @danielkumor ): “Causal Imitation Learning with Unobserved Confounders,” & stop by our oral presentation later today (09:15 EST).

Tweet media one

Tweet media two

Tweet media three

Tweet media four

1

18

106

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

Neural causal models were introduced in + the 1st general algorithm to perform causal inference entirely *inside* a neural net! There is indeed a tradeoff bt symb. & OPT-based methods for ID, elaborated in App C4, p 43; see also FAQ-Q11 @ylecun @GaryMarcus

Tweet media one

Tweet media two

Tweet media three

Tweet media four

@yudapearl

Judea Pearl

2 years

@johnmark_taylor Bottom line: Identification = Symbolic, Estimation = Deep Learning. More details: Also:

4

6

39

2

22

109

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

The slides of the WHY-19 talks are now available on our website, see: . Thank you, @yudapearl @PHuenermund @murat_kocaogIu @tdietterich @tobigerstenberg @MariaGlymour

1

32

105

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

The call for papers for the Journal of Causal Inference's special issue on the data-fusion challenge -- combining observational and experimental data -- is out! I will be co-editing it with @mark_vdlaan . Please, consider submitting your best work:

1

37

105

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Our paper "Causal Inference and Data-Fusion in Econometrics” is finally out, . We discuss recent results in Causal AI in the context of Econometrics. It has been a pleasure to work w/ @PHuenermund , who wrote a nice tweet-thread explaining the paper (below).

@PHuenermund

Paul Hünermund

4 years

Finally, @eliasbareinboim and I are able to share a working paper that we've been working on for quite a while now: "Causal Inference and Data-Fusion in Econometrics" 🚨 #DAG #CausalInference #Econometrics #DataScience #MachineLearning #AI 1/

Tweet media one

9

180

572

1

24

103

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

We are developing a tool called ‘Fusion’ that is fully compatible w/ the #BookOfWHY , following the discussion in Ch. 10. Fusion offers an easy-to-use way of doing causal inference & fusion from 1st principles, incl. do-calc. If interested, subscribe here .

Tweet card media

Causal Fusion Beta Testing

docs.google.com

@catala_fr

Francois Catala

4 years

@yudapearl @wiredmau5 @yudapearl What do you think about DoWhy and CausalImpact? Are there any tools that you would recommend for when you are ready to put what you have read into practice? Thanks! #Bookofwhy

0

0

0

3

29

98

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

I will be talking today in the MSR's Frontiers in Machine Learning 2020 -- "On the Causal Foundations of Artificial Intelligence (Explainability & Decision-Making)" -- . If you are attending the event, I would be glad to chat more. @amt_shrma @emrek

Tweet card media

Frontiers in Machine Learning 2020 - Microsoft Research

Contact Us: If you have questions about this event, please email us at [email protected] Opens in a new tab

www.microsoft.com

1

4

85

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

If you are around Palo Alto, I’ll be speaking at the Stanford Business School ( @StanfordGSB ) about “Causal Data Science’, i.e., how to make transparent & principled causal inferences from data. When: Mon Oct/7, 1:10 - 2:30pm; Where: Rm G101, Gunn Bldg. @StanfordMed @StanfordEng

4

3

85

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

I'll be talking about causal inference & the data-fusion next week at Harvard, see the link below. If you are around, consider stopping by. I'll be covering topics such as transportability & generalizability as discussed in this PNAS paper , See you!

Causal inference and the data-fusion problem | Proceedings of the National Academy of Sciences

We review concepts, principles, and tools that unify current approaches to causal analysis and attend to new challenges presented by big data. In p...

@HiTSatHarvard

HiTS @ Harvard

5 years

@HiTSatHarvard Seminar Series: "Causal Inference and Fusion" Elias Bareinboim, PhD @eliasbareinboim Purdue University, Dept of Computer Science @PurdueCS Thursday, March 21, 2019 10:00-11:00 Harvard Medical School, WAB563 #HiTSSeminar

2

2

11

4

17

86

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

"Purely evidence-based policy doesn’t exist" - Nice essay by UChic-Econ, Lars Peter Hansen ( @UncertainLars ); similar to points repeatedly raised by J. Ioannidis ( ) & @yudapearl ( #BookOfWhy ) in other fronts. @PHuenermund @analisereal

Tweet card media

Evidence-based medicine has been hijacked: a report to David Sackett - PubMed

This is a confession building on a conversation with David Sackett in 2004 when I shared with him some personal adventures in evidence-based medicine (EBM), the movement that he had spearheaded. The...

pubmed.ncbi.nlm.nih.gov

0

29

84

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

Regardless of RL, which has interesting connections w/ CI (), Deep & Causal modes of reasoning are connected in a fundamental way, as recently discovered: . I believe this starts addressing Judea's concerns, as quoted in footnote 4.

@roydanroy

Dan Roy

2 years

@yudapearl I think the DL people working on RL were too busy to respond?

3

1

11

1

17

81

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

If you are in Boston, I’ll be speaking at the @MIT graphical models workshop on “Causal Data Science’. When: Today 9:30 am; Where: MIT Bldg 2 (Math). I'll be around for a few hours after the talk, drop me a line if you are interested in talking more about CDS. @mitidss @MIT_CSAIL

1

9

81

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

I'll be talking about causal inference & data-fusion at MIT next week; see details . If you are around, consider stopping by. I'll be covering topics such as generalizability & transportability, following our PNAS paper … See you!

Causal inference and the data-fusion problem | Proceedings of the National Academy of Sciences

We review concepts, principles, and tools that unify current approaches to causal analysis and attend to new challenges presented by big data. In p...

@mitidss

MIT IDSS

2 years

How can we handle the biases that emerge when piecing together multiple datasets collected under heterogeneous conditions? Join us next Monday (May 2) for an IDSS Distinguished Seminar with @eliasbareinboim on Causal Inference and Data Fusion.

1

7

22

5

18

79

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

Agree, my take on generalization related to causal knowledge: . The foundations are pretty stable but there have been some newer results since I wrote this around 2014. Thanks @yudapearl for the inspiration & partnership. Summary:

Tweet media one

@GaryMarcus

Gary Marcus

2 years

🔔Virtually every important question in AI revolves around the scope and limits of generalization—yet there is little consensus about what generalization even means or how to test it. Check out @_dieuwke et al’s new synthesis, @genbench , to get some much-needed clarity.

7

15

80

2

26

78

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Nice blog post () by @an1lam discussing some common misconceptions about Causal Inference & summarizing some of what he learned in my course this Spring. It was nice having you around, Stephen, thanks!

1

13

72

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

How causality can further advance data science and AI #bookofwy

Tweet card media

Why every data scientist shall read “The Book of Why” by Judea Pearl

How causality can empower data science/ AI

towardsdatascience.com

0

13

70

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

@dustinvtran @zacharylipton @CsabaSzepesvari @GaryMarcus @earnmyturns @yudapearl @ShalitUri Sorry for the self-citation, but a lot is happening - e.g., off-policy method thr. causal reason.: ; Where to intervene? ; Counterfactual bandits: ; Learn causal model from do-dist():

1

5

70

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

If you are at ICML, check our paper (w/ @YonghanJung & Jin Tian) on the first family of Double-ML estimators for any identifiable effect computable from observational data (i.e., equiv. class of DAGs): When: July/23, 12-2 am (EST),

Estimating Identifiable Causal Effects on Markov Equivalence Class through Double Machine Learning

General methods have been developed for estimating causal effects from observational data under causal assumptions encoded in the form of a causal graph. Mos...

proceedings.mlr.press

1

11

71

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

1/ The words “model” & “diagram” seem to be overloaded & shouldn't be conflated, which we clarified in our new chapter (). i) Model (top left, Fig 1.2) = A structural causal model (SCM) is a collection of mechanisms underlying a system & it’s always there.

Tweet media one

@JPedro_Monteiro

JP Monteiro

@JPedro_Monteiro

4 years

@yudapearl One thing I missed in "The Book of Why" was on how causal models come to be. Most of the discussion in the book already assumed some kind of model/diagram. Many "curve-fitting" ML practitioners I talked to struggle especially on this part and are therefore skeptical about it.

3

1

10

1

13

70

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

8 years

My paper with Pearl on causality & big data is available online - Causal inference and the data-fusion problem #PNAS

0

8

68

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

One way of attaching causal meaning in regression is thr Instrumental Variables (Wright, 1928). In a new paper w/ @analisereal @danielkumor , we unify the literature on linear identification & develop a method for finding Instrumental Cutsets (general. IV)

Tweet media one

0

20

67

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

Carlos Cinelli ( @analisereaal ) just gave a wonderful talk about external validity, transportability, and generalizability in causal inference, which is at the core of what CI is all about. The slides are available here: .

1

12

67

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 months

Thanks for sharing your thoughts, Amit. Recall, just to add some clarity in terms of context, my comment regarding @ylecun & @yudapearl 's posts is neither about generative nor about deep learning versus causal; those are pacified issues in the literature. In other words, we now…

Tweet media one

Tweet media two

Tweet media three

@amt_shrma

Amit Sharma

2 months

So from this perspective, I see the value of both @ylecun and @eliasbareinboim viewpoints. True that we cannot learn causal agents simply from observing the world, since observed data is usually confounded.

1

0

14

4

17

67

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

For those who are curious, this is the paper the cute baby is learning about, "Estimating Causal Effects Using Weighting-Based Estimators" (). The task is to allow the estimation of identifiable expressions that go beyond the usual backdoor/IPW cases.

@yudapearl

Judea Pearl

4 years

My My!! He got it right!! The most inspiring photo I've seen on Twitter in the past year!!

0

4

156

1

11

64

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

If you are interested in a postdoc position in Causal AI (my lab), consider applying to the CRA fellowship (link below)!

0

12

62

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

Thanks for sharing this. Somewhat curious perspective by on causality. Apparently, he believes that causality is just one among "hundreds of different things that deep learning doesn't do well" & his example of another is "explainability"! Thoughts?

Tweet media one

@yudapearl

Judea Pearl

2 years

I bring this up, not as indictment of ML leaders but as an enticement of their students. These "goodies" are essentially solved using causal modeling, a fact that should jolt every curious student into asking: "HOW?" and into exploring the Laws and tools of Causal Inference.

4

4

52

6

15

63

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

23 days

Thanks for writing the piece, @erichorvitz ! I think "causal inference" should be put front and center since, as part of the AI community, we could help provide sound foundations for many challenges (including safety, equity, robustness, transparency, and understanding). Happy to…

@erichorvitz

Eric Horvitz

24 days

"Now, Later, and Lasting: 10 Priorities for AI Research, Policy, and Practice," forthcoming in CACM, with @conitzer @SheilaMcIlraith @PeterStone_TX #AI100 @CACMmag @TheOfficialACM @StanfordHAI @PartnershipAI @WHOSTP @CarnegieMellon @UofT @CIFAR_News

2

18

61

2

13

63

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Insightful note by investor @RayDalio on the necessity of understanding cause-effect relations to make robust decision-making & the insufficiency of machine learning, (~3 mins): . We do have a language & tools to encode this understanding today! @yudapearl

Tweet card media

What coronavirus means for the global economy | Ray Dalio

Visit http://TED.com to get our entire library of TED Talks, transcripts, translations, personalized talk recommendations and more."I'm a capitalist. I belie...

www.youtube.com

2

17

59

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

1/3 This seems to be a nice extension of the work causal-neural connection () for GNNs, but I haven't fully read it (congrats, @kerstingAIML & team!). Still, we answered precisely this question in Q11 in the FAQ (p. 52), also attached for your convenience.

Tweet media one

@yudapearl

Judea Pearl

3 years

2/ is: "What is gained by this translation?" Namely, why not conduct the analysis in the SCM framework and let GNN take over the estimation part only, as is done here An explicit characterization of the gain would be helpful.

2

3

38

4

11

53

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

I believe what you are looking for is called "causal inference" (CI) - every problem I found in RL is solved from 1st principles by tools developed within the CI framework. For a few cataloged examples, see , ,

@togelius

Julian Togelius

4 years

I don't have any beautiful new framework to replace RL with. But I hope that during my lifetime someone will come up with one. My guess is that it will not come out of current formalisms, such as the POMDP framework, and that most of them will be pointless after the revolution.

12

5

81

2

7

51

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 months

@ylecun @yudapearl > "You don't necessarily need RL to train a causal world model." (...) "Only SSL from off-line observation data and planning." If I understand what you are saying, in addition to problems of scale that we can leave aside for now, your proposal assumes that the way the offline…

2

3

51

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

Thanks for your interest, @haschyle , I just posted the slides here: . The talk was recorded and I expect the video should be available sometime in the near future, stay tuned. @StanfordGSB

@haschyle

Achille

5 years

@eliasbareinboim @StanfordGSB @StanfordMed @StanfordEng Will the talk be recorded/slides available?

0

0

0

0

1

50

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Can 'Causal Data Science' be helpful for the future of the Amazon Forest? This idea has been first articulated by @juanccas ( @wef , @EBPgenome ) in a thoughtful, new piece that just appeared at the Exponential View, @PierreGentine @runge_jakob @yudapearl

Tweet card media

🔮🤩 The inclusive bioeconomy; the future of Amazon; synthetic biology, and fires from the space++...

Could a new paradigm protect the environment and advance economies?

www.exponentialview.co

0

8

50

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

Nice rebuttal to Sutton's note by @wellingmax . The folks doing causal inference may be excited to read the note since we learn case after case that there's no such thing as model-blind causality. Certainly worth reading to see how causality fits the larger conversation in ML.

@wellingmax

Max Welling

5 years

I wrote this response to Rich Sutton's "The Bitter Lesson"

18

232

699

1

9

46

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

8/8 If anyone is in the NY area & wants to grab a coffee, drop me a line. We should have a friendly & intellectually stimulating environment to receive scholars interested in learning more about CI, doing research in the field, or using CI methods in their own research. See you!

8

0

46

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

I'll be talking about causal inference and data-fusion this Monday at Columbia. If you are around, consider attending.

@ColumbiaCompSci

ColumbiaCompSci

@ColumbiaCompSci

5 years

@eliasbareinboim will present "Causal Data Science: A general framework for data fusion and causal inference" as part of the Distinguished Lecture Series on Monday, April 1 at 11:40am in CSB 451. #DataScience #causalinference

Tweet media one

1

6

10

4

8

44

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

1/2 I don't disagree but my feeling from interactions at NeurIPS this year is that there's a deeper phenomenon going on. Folks want to do CI since they feel it's important, incl. because of your new book, but they haven't spent the time & energy learning what CI is really about.

@yudapearl

Judea Pearl

4 years

@ngutten @jsusskin Interesting viewpoint. Trouble is, regression is such a small part of CI that to say "To do CI with regression requires an extra step" is almost like saying: "To do CI with algebra requires an extra step". Its better to say "Do CI first, add NN if needed" @ylecun #Bookofwhy

1

0

8

3

1

42

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

2/6 Tue 11:30 am (EST) “The Causal-Neural Connection: Expressiveness, Learnability, and Inference”, with Kevin Xia, Kai-Zhan Lee, and Yoshua Bengio. Link: .

Tweet media one

Tweet media two

Tweet media three

Tweet media four

1

5

39

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

1 month

Hi Judea, I just learned about Danny from your tweet. His work is really groundbreaking and he is a unique human being, very generous, humble, and curious. I have some memories and will share one. We first met in person when I was interviewing here at Columbia about 5 years ago.…

@yudapearl

Judea Pearl

1 month

Israeli American economist and Nobel laureate Daniel Kahneman dies at 90

20

66

438

0

1

41

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 months

@ylecun @yudapearl Yann, what is the difference between what you call "causal prediction" and computing next-state probabilities in a typical reinforcement learning (RL) setting? As we elaborated through our work on Causal RL (), there are exciting things coming from RL, but…

3

5

39

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

1/4 Broadly, because estimating a cond. prob. P(Y | X) is easier than a joint one, P(Y, X, Z), when X & Z are high-dimensional. Indeed, this observation led to (non-causal) graphical models in the 1980s, including Bayesian nets (ie, non-causal DAGs), Markov random fields & so on.

@deaneckles

Dean Eckles

3 years

@VC31415 @Chris_Auld @Sander_vdLinden @GordPennycook @AdamBerinsky @lkfazio @DG_Rand What would the point of such a factorization be? Why would one, say, prefer estimating a conditional expectation to an unconditional one?

1

0

2

1

6

41

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

1 year

For the folks following the thread, the disentanglement of the assumptions and conclusions is essential in data science. After months of discussions & going around, Ivan’s clean & crisp analysis sheds light on a pervasive issue of transportability. Thanks for sharing this, Ivan.

@ildiazm

Iván Díaz

1 year

In this post and multiple others, @f2harrell says that a generally correct way to analyze RCTs with binary outcomes is to fit a logistic regression. One of the main claims is that this leads to "one-number" effect measures that are "highly transportable".

11

22

156

1

5

40

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

No, but my UAI tutorial this year, which is a longer version of this talk, was recorded and should be available sometime soon... Also, our CRL survey should be made available sometime soon, I hope. Thank you for your interest, I'll keep you guys posted.

@zacharylipton

Zachary Lipton

5 years

@eliasbareinboim @djhsu Will it be recorded?

0

0

3

0

2

40

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

6 years

Review NYT: ‘The Book of Why’ Examines the Science of Cause and Effect #causalinferencebook

Review: ‘The Book of Why’ Examines the Science of Cause and Effect (Published 2018)

A longtime U.C.L.A. professor delivers not only a valuable lesson on the history of ideas but provides the conceptual tools needed to judge just what big data can and cannot deliver.

www.nytimes.com

2

17

39

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

1/12 Brdly spk, assuming one can infer the effect of an intervention when experimental data matching exactly this intervention is available is not surprising, causally-speaking. Why? There's no cross-rung inference, but a typical function approx exercise. This is Judea's main...

@kudkudakpl

Stanisław Jastrzębski

4 years

@yudapearl @GaryMarcus @moultano What precisely do you mean by "mathematical impossibility"? Causal graph can be inferred if observations are sampled using different random interventions. Such inference can be done using DL . Then CI can be done on top using another DL model

1

1

5

3

2

39

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

What is the Purpose of Statistical Modelling? (by David Hand) -- . Nice piece by David Hand that called my attention since, while only implicitly, it seems to suggest a connection of mainstream Stats with the Ladder of Causation.

Tweet card media

What Is the Purpose of Statistical Modeling? · Issue 1.1, Summer 2019

hdsr.mitpress.mit.edu

2

7

36

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

We have an example written to answer this issue; see p. 28, example 10 (), and summarized in the diagram below. We can have different Bayes nets generating the same obs. data (P(V) ) & that naively would entail different do-distributions P(Y | do(X)).

Tweet media one

@VC31415

Victor Chernozhukov #peace 🇺🇦

3 years

@julianschuess But people do this all of the time. They fit a statistical model to the data (a markov net over a DAG), and do "causal" calculations. Is there an interpretation, especially given that there is fundamentally no way of verifying from the data if the model is causal or not?

3

0

7

0

8

37

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

Thank you, Judea! We will keep your offer open for an indefinite duration. :)

@yudapearl

Judea Pearl

3 years

Speaking of advertising for a job or a postdoc, here is one from @eliasbareinboim I do not know what could be more "innovative" than research on Causal AI. If I was qualified I would go for it.

1

4

29

1

2

36

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Thanks! Lang. L1 (association), L2 (intervention), L3 (counterfactual) are increasingly complex & higher-order lang. include the lower ones. Sec 1.3 formalizes the expressiveness issue; Theorem 1 (pp. 22) shows the containment is strict; for intuition, see examples 7-9 (~p. 27).

@VC31415

Victor Chernozhukov #peace 🇺🇦

4 years

@PHuenermund @yudapearl @eliasbareinboim Great paper. It is not clear to me why L3 questions are separated from L2. It seems that L2 is a special instance of L3. L2 is done using do() notation, but this is abandoned when doing L3.

4

1

13

1

1

35

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

The goal shouldn't be "good in-distribution generalization" but extracting the proper invariances to work in the real world, which distribution rarely matches the sample (chang. conditions). In causality, this task is known as statistical transportability,

@nasim_rahaman

Nasim Rahaman

4 years

@yudapearl If I were to think of a reason, it’d be that a good chunk of ML is about getting good in-distribution generalization (yes, “curve-fitting”). The lack of a sound causal treatment doesn’t bite that hard in this setting.

0

0

2

1

1

34

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

6 years

Following @causalinf request, here are some of @yudapearl personal memories from his early days, which were taken out of the #bookofwhy : . It's still unpublished material but authorized for your enjoyment. More to come. #bookofwhy

1

6

34

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Everyone wants data-driven smth, sure, but Judea's perspective is technical, rooted in the Causal Hierarchy Theorem (CHT) (p. 22, ). The CHT says that for almost any causal model, the layers of the hierarchy do not collapse (examples 7-9 are illustrative).

@wellingmax

Max Welling

4 years

@yudapearl It’s unlikely that data-driven ML will be replaced. Rather it will be augmented by causal modeling. And I bet that even causal modeling will eventually become data-driven itself in the form of causal generative models.

2

19

181

2

7

33

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

@tdietterich @hardmaru I would just add that even though not all causal invariances are learnable from interactions, some of them are. I think the most promising setting is reinforcement learning, where interventional capabilities are available (e.g., , ).

2

2

33

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

2/5 Tue 4:15 pm/6:30 pm (EST) “Partial Counterfactual Identification from Observational and Experimental Data”, with Justin Zhang & Jin Tian. Link: .

Tweet media one

Tweet media two

Tweet media three

1

6

33

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

Thanks for sharing your thoughts, K. The universality of NNs does NOT help causal reasoning. In general, it's not that a task would be "harder," but impossible to solve. I am sharing below a recent paper that investigates this issue from first principles,

Tweet media one

Tweet media two

Tweet media three

@sirbayes

Kevin Patrick Murphy

2 years

Some random musings on how the whole "AI scaling" debate. It is clear that neural nets can model any predictive distribution p(x_future|x_past) given enough parameters and data, since they are universal approximators;

13

147

893

0

5

31

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 months

1/5 Happy to share some of our latest work that will be presented this week at NeurIPS in New Orleans! The authors would be delighted to see you at the poster session and talk more about our current work and future challenges! Tue 6:15 pm (Poster Session 2) "Causal discovery…

Tweet media one

Tweet media two

Tweet media three

Tweet media four

0

5

33

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 months

@ccaballeroh10 @yudapearl @ylecun @sirbayes The terminology game is indeed complicated and usually arises when we attempt to make interdisciplinary claims, but point well taken. To focus on the substance, my suggestion for those interested in learning more is to first: (1) understand what control theory has accomplished,…

1

3

31

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

For those interested in causal inference & around NYC, I'll be talking about causal data science & modeling today (Nov 21) at CUIMC. Time: 1:00 pm - 2:30 pm. Location: Hammer Health Sciences Building, 701 West 168th Street, Room # LL-106 @ColumbiaMSPH @ColumbiaMed @DSI_Columbia

3

2

32

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

1 year

@ildiazm It's surprising to hear @f2harrell saying that, it seems the opposite of how things evolved. I am attaching an intro written in Stat Sci ~10yr ago on threads vs. assumptions in the ctx of RCT's external validity, where transportability theory started.

Tweet media one

1

4

32

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

4/5 Tue 6:30 pm (EST) “On Measuring Causal Contributions via do-Interventions”, with @YonghanJung , D. Janzing, J. Tian, S. Kasiviswanathan. Link: .

Tweet media one

Tweet media two

Tweet media three

Tweet media four

2

4

31

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

5/5 Mon 3:30 pm (EST) I will be presenting with Drago a tutorial on “Causal Fairness Analysis”. Links: and

Tweet media one

Tweet media two

Tweet media three

Tweet media four

1

3

30

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 years

The wisdom in this line is quite remarkable & largely unknown by most folks outside the field (& some reviewers inside :)). My 2 cents: this is one of the first instances where the requirement of having a model was relaxed after understanding it seriously:

@yudapearl

Judea Pearl

5 years

2/2 Along a similar vein, I was asked to retweet the last line of my slides in the Why-19 symposium . Gladly; it reads: "Only by taking models seriously we can learn when they are not needed." And I still vow for it. #Bookofwhy

2

8

38

1

0

30

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

1/3 From Hume (footnote 2, ): “Nature has kept us at a great distance from all her secrets, & has afforded only the knowledge of a few superficial qualities of objects; while she conceals from us those powers & principles, on which...

@hardmaru

hardmaru

4 years

I think an AI system's ability to perform tasks like casual inference and symbolic processing will eventually be learned or evolved as an emergent property that just happens to be useful for its environment. Not be something that is formally defined and hand-engineered by humans.

41

82

640

1

5

28

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

Scientific progress happens but at a slower pace than it should by any reasonable standard (see, eg, ). I volunteer to help, perhaps talking w/ some of the agencies leaders? Reviewers are extra conservative, perhaps need enlightened leadership? @tdietterich

Planck's principle - Wikipedia

en.wikipedia.org

@yudapearl

Judea Pearl

4 years

Our well-intentioned funding agencies are partly responsible for this mis-balanced mis-investment. They all want & fund "explainable AI" "robust AI" "life-long learning" etc etc. forgetting that, once projects are run by traditionally-trained PI's they end up "more of the same."

0

6

36

4

0

27

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 months

Hi Damien, thanks for raising the issue. I am not sure I fully buy into the "camps" comparison, with all due respect, and I would add that most folks I know who are doing causal inference research are not focused on the worst-case scenario but on systematic "understanding," which…

@DamienTeney

Damien Teney

2 months

My meta impression on "ML optimists" & "causal reasoning" camps: - one tends to focus on empirical results about what can be done (with observational data, scaling...) - the other on principled reasons about what can't be done (worst case scenarios) We need both points of view.🙏

0

0

6

4

6

22

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

6/6 Wed 7:30 pm (EST) “Nested Counterfactual Identification from Arbitrary Surrogate Experiments”, with Juan Correa & @sanghack . Link: .

Tweet media one

Tweet media two

Tweet media three

Tweet media four

3

6

28

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

Nice & informative introductory post on Causal Data Fusion! Congrats, @PHuenermund !

@PHuenermund

Paul Hünermund

3 years

New post on @causal_science explaining the data fusion paradigm – an integrated and fully automatable framework from problem specification, to identification, to estimation. #CausalInference @yudapearl @eliasbareinboim #Causality #DataScience #ML #AI

2

39

146

0

6

28

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

The parameters of the SCM are almost never identifiable from obs. data, check the Causal Hierarchy Thm (p. 22) & examples 7-9 (p. 24-26, ). Still, estimating effects using DAGs as proxies for the SCMs makes total sense, e.g., see .

Tweet media one

@jwalkrunski

Jeff Walker

3 years

@eliasbareinboim @deaneckles @pablogerbas I missed the scope of this statement "so estimating/fitting the parameters makes no sense" -- is this specific a paper or do you mean it never makes sense to estimate parameters using a DAG?

0

0

1

0

6

27

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

4 years

@yudapearl 's talk at CIFAR Machines & Brains workshop -- "Data versus Science: Contesting the Soul of Data-Science": . (Due to some Zoom issues, the synchronization of the video & slides was not perfect, apologizes!!) @bschoelkopf @MILAMontreal @CIFAR_News

0

8

27

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

5 months

5/5 Thr 11:45 am (Poster Session 5) "A Causal Framework for Decomposing Spurious Variations" (joint work with Drago Plecko) In this paper, we study confounded variations, which are a fundamental aspect of constructing explanations and may help to answer questions ranging from…

Tweet media one

Tweet media two

Tweet media three

Tweet media four

1

6

27

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

6 months

Hi @ylecun , thank you for bringing attention to this concerning picture of the current situation. From my observation, many academics, myself included, might not be so closely monitoring the issue, and as a result, might be unaware that certain individuals or companies are…

@ylecun

Yann LeCun

6 months

@tegmark @RishiSunak @vonderleyen Altman, Hassabis, and Amodei are the ones doing massive corporate lobbying at the moment. They are the ones who are attempting to perform a regulatory capture of the AI industry. You, Geoff, and Yoshua are giving ammunition to those who are lobbying for a ban on open AI R&D. If…

317

1K

6K

0

2

26

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

1/2 I would phrase it more precisely - NCM is a special class of SCMs used as proxies of the true SCM & where the functions are neural nets. IF one wants to solve an identification task, eg, yes, a search over the NCM space would be entailed. App. C4 (p 42) discusses this point.

@analisereal

Análise Real

3 years

@eliasbareinboim If I understand correctly, the NCM approach is a numerical search over the space of neural networks?

0

0

0

2

3

26

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

3 years

2/6 Wed 12:00 pm (EST) “Causal Discovery from Soft Interventions with Unknown Targets: Characterization and Learning”, with Amin Jaber, @murat_kocaoglu_ , Karthikeyan Shanmugam. Link: .

Tweet media one

1

2

25

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 months

Hey @andrewgwils , thanks for sharing your thoughts. I am not a radiology expert, but to address a broader, related point -- isn't the success of these methods in this setting due to the huge amount of annotated data available, coming from real, qualified radiologists? As far as I…

@andrewgwils

Andrew Gordon Wilson

2 months

@MelMitchell1 @geoffreyhinton The whole point is he wasn't trying to be precise about the exact timescale. It's not 10 vs 5. It was more: "we should start changing the way we train radiologists, and those in professions where we can collect ample data and be greatly assisted by machine learning".

2

0

14

1

5

26

@eliasbareinboim

Elias Bareinboim

@eliasbareinboim

2 years

3/6 Tue 11:30 am (EST) + Oral Fri 7:40 pm (EST) “Sequential Causal Imitation Learning with Unobserved Confounders”, with @danielkumor & Junzhe Zhang. Link: .

Tweet media one

Tweet media two

Tweet media three

1

5

23