(2023) Causality, LLMs & Abstractions || Matej Zečević || Causal Bandits Ep. 000

Recorded on Aug 14, 2023 in Frankfurt, Germany.

Audio version available on selected platforms: https://causalbanditspodcast.buzzsprout.com

*Are Large Language Models (LLMs) causal?*

Some researchers have shown that advanced models like GPT-4 can perform very well on certain causal benchmarks.

At the same time, from the theoretical point of view it’s highly unlikely that these models can learn causal structures.

Is it possible that large language models are not causal, but talk causality?

In our conversation we explore this question from the point of view of the formalism proposed by Matej and his colleagues in their “Causal Parrots” paper.

We also discuss Matej’s journey from the dream of becoming a hacker to a successful AI and then causality researcher.

Ready to dive in?

——————————————————————————————————

*About The Guest*
Matej Zečević is a researcher and a PhD candidate at TU Darmstadt. His work is focused on the intersection of causality, machine learning (ML) and other branches of artificial intelligence (AI). Learn more about Matej and his research here: https://www.matej-zecevic.de/
Matej on Twiter/X: https://twitter.com/matej_zecevic/

*About The Host*
Aleksander (Alex) Molak is an independent machine learning researcher, educator, entrepreneur and a best-selling author in the area of causality (https://amzn.to/3QhsRz4).
Connect with Alex:
– Alex on the Internet: https://bit.ly/aleksander-molak

*Links*
Events
– Causality Discussion Group (https://discuss.causality.link/)
– Eastern European Machine Learning Summer School (https://www.eeml.eu/home)

Videos
– Prof. Moritz Helmstaedter on connectomics (https://www.youtube.com/watch?v=uNyDSx14yIQ)

Books
– Molak (2023) – Causal Inference and Discovery in Python (https://amzn.to/3QhsRz4)
– Pearl & Mackenzie (2019) – The Book of Why (https://amzn.to/40eRHUV)
– Peters et al. (2017) – Elements of Causal Inference: Foundations and Learning Algorithms (https://amzn.to/3sIfG2t)
– Stanley (2015) – Why Greatness Cannot Be Planned: The Myth of the Objective (https://amzn.to/3SJdS3D)

Papers
– Beckers (2019) – Abstracting causal models (https://ojs.aaai.org/index.php/AAAI/article/download/4117/3995)
– Kaddour et al. (2022) – Causal Machine Learning: A Survey and Open Problems (https://arxiv.org/pdf/2206.15475.pdf)
– Kiciman et al. (2023) – Causal reasoning and large language models: Opening a new frontier for causality (https://arxiv.org/pdf/2305.00050.pdf)
– Mooij et al. (2016) – Distinguishing cause from effect using observational data: methods and benchmarks (https://jmlr.org/papers/volume17/14-518/14-518.pdf)
– Rubenstein et al. (2017) – Causal Consistency of Structural Equation Models (https://auai.org/uai2017/proceedings/papers/11.pdf)
– Zečević et al. (2021) – Relating Graph Neural Networks to Structural Causal Models (https://arxiv.org/abs/2109.04173)
– Zečević, Willig et al (2023) – Causal Parrots: Large Language Models May Talk Causality But Are Not Causal (https://openreview.net/pdf?id=tv46tCzs83)
– Zečević et al. (2021) – Causal Explanations of Structural Causal Models (https://arxiv.org/abs/2110.02395)
– Zečević et al (2023) – Not All Causal Inference is the Same (https://openreview.net/pdf?id=ySWQ6eXAKp)

*Causal Bandits Team*
Project Coordinator: Taiba Malik (https://www.instagram.com/taibasplay/)
Video Editors: Maaz Ali, Aleksander Molak

*Action*
Connect on LinkedIn: https://www.linkedin.com/in/aleksandermolak/
Join Causal Python Weekly: https://causalpython.io
Causal Bandits: https://causalbanditspodcast.com
The Causal Book: https://amzn.to/3QhsRz4

#machinelearning #causalai #causalinference #causality #llm

It’s a double-edged source because there’s a lot of wrong intuition that we have about causation hey causal Bandits welcome to the causal Bandits podcast the best podcast on causality and machine learning on the internet today we’re traveling to Frankfurt to meet our guest he worked in robotics and wanted

To become a hacker he’s a passionate person and a dedicated researcher with interest in AI philosophy and Neuroscience he co-authored papers at the intersection of causality and large language models graph neural networks and more a firm believer in human potential and a passionate Community leader ladies and gentlemen Mr

Maich let me pass it to your host Alex moac hi mat how are you today I’m doing fine and how are you fine yeah beautiful I’m also good a little bit tired after the trip as always uh but that’s great I’m super excited about our conversation

Today me too uh I’m looking very much forward to this uh episode today you’re a very successful uh researcher and a very successful Community leader now you co organized uh np’s Cal workshop last year uh you published or call for papers with people like deep Minds p valovich

Uh UD talked about your work on on on Twitter how all of this has started for you okay this this is a lot to digest um I guess I’ll start with the first so uh thank you for the compliment I don’t see myself that way um I’m trying to be so

If the perception is is is coming to to this point then I guess I’m doing something right um but in general I’d say that all of this started really we were talking about this uh just recently uh with with the book of why Pearl’s book which sparked the flame of interest

In here of course I was already studying AI uh in my computer science studies at T damad um but yeah that’s that’s where it kicked off and then from there is I guess just a culmination of tools and and and tips and tricks and all the amazing people who have supported me all

Along the way M uh which has led to this point that I can use all of these things and and I guess things like community and so on have always been important to me so you talked about book of why the book of why which means I understand you

Focus on the on the causal part of of the equation um but causality was not the first Topic in in machine learning AI or computer science for you right correct so I actually started my um computer science studies my undergrad in t dad in my bachelor um and and and

Wanted to do computer security it security and then really quickly through uh Professor Yan Peters who is a big name in robotics and reinforcement learning uh convinced me otherwise uh very very convincingly and um this way I started in Rob robotics a little bit then it went into a bit of

Neuroscience all kinds of topics related to intelligence um before then when I was in my masters I I I suddenly stumbled upon this book right and you know a long tradition of Statistics just got overwhelmed and um and yeah that’s that’s how we got here so what’s the

Common denominator for for all of this security uh Robotics and then and then finally causality is there something that some question that you had some more basic question that you are trying to to answer find some answers uh looking into those those three fields or maybe more Fields so if I take security

For instance that’s an interesting question then then I’d say it was really just about the coolness factor of you know um uh you know adolescent uh who is trying to be cool trying to be a hacker like they portray in the movies um I guess with intelligence is a bit similar

But uh with intelligence it really was more the foundational question trying to understand oneself right giving oneself meaning in a sense right might be broad speaking now but but still this is I guess deep down what motivated me um but if you take now the subtopics in regards

Say uh reinforcement learning uh I mean related fields like Neuroscience cognitive science continual learning all these different aspects of of learning right statistical learning theory um then I think the the common thing is really trying to to figure out what intelligence is get it you know operationalized um but coming from

Different perspectives right so I I always felt that reinforcement learning was closer to robotics while uh you know Neuroscience cognitive science they are closer to the human but all have the same ultimate goal so so there was a personal question behind all of this in a in a sense

Yeah I guess so um as you said uh I’m actually yeah genuinely passionate about everything and and I guess that’s just stems from my from my person genetically e ethically and also ethnologically and all these aspects right which which make a person so um I’ve always been very passionate about all these different

Things and uh I guess you know the the Natural Curiosity of a scientist maybe yeah that’s really what sped for me and given that intelligence is by no means resolved it’s not really even defined right we all have some kind of intuition I guess for me the best definition is

Actually that um we are our own example of it right so so whatever we are and and that’s how how I also go about causal inference right that that that’s the key argument for me why it’s necessary not that it’s sufficient but it’s necessary suff necessary but not

Necessarily sufficient for for what uh for intelligence for intelligence yes so reasoning causally would be would be this necessary but not a sufficient that’s my personal belief right it could also be sufficient I guess if you ask someone like UDA he would say probably it’s also sufficient um to me it’s

Definitely necessary but not necessarily sufficient because there’s all these different topics which are kind of hard to grasp like emotions and things like that MH regarding your curiosity what fascinates you the most or what do you find the most promising Direction in in causal research today

That’s a difficult one so I always think of it as these two strands right so we as you know I also organized this discussion group by the way please check it out if you haven’t so far if you’re interested we’ll give you a link in the

In the uh show notes it’s the weekly coal discussion group and and we had a session by uh Mel vinst he’s from I think it was Luc University of Lubec in the north of Germany they do some amazing work shout out to Mel and um he was actually uh talking about this that

There’s like causal Discovery and then there’s this inference part right where the inference part is more about modeling assumption about you know having sound conclusions based of what you have and what you have is usually then assumptions on the graph while causal Discovery is uh the problem of

Getting that graph in the first place from data right it’s it’s more machine learning in a way yeah and um these are kind of like the two two main strands and so research is happening there and super important uh I think what is missing though is kind of the the bridge

Between the two and and also what is missing and what I’m trying to do with my own research actually is um more philosophical in nature less scientific more philosophical in the sense that uh have we asked all the questions right like broadening The Horizon looking outside of the box right and and yeah

Just just raising questions we might have not asked before and obviously that goes with you know questions that might not be of any relevance but well that’s where you try right and and maybe something will actually be very relevant yeah so I think it’s about interfaces about broadening The Horizon these are

Missing things um and apart from that I think especially in machine learning now it’s it’s very much about this discovery part right like learning representations and things like that um to me actually a little bit what is missing still and which I personally find very important is abstractions so there has been this

Work by S beas also shout out to you s um and and romstein and and all others who pioneered in this direction um which is kind of analytical philosophical so very Theory grounded um and there’s first attempts so also the group from moris of in in in Vienna they are really

Pushing for this um but still it’s it’s it’s kind of a bit less yeah also what I would maybe mention as a as a as a final Point here is um the connection to logic so as you said we organized this Nur workshop last year 2022 and it was titled uh neuroc causal

And symbolic AI so neuro was this part for neural networks deep learning all the modern stuff causal for causality and symbolic for all the well logic based work symbolic world as we say the GOI the good oldfashioned AI um and and and neuros symbolic stuff which is kind

Of the intersection of of of these two worlds to talk to cality because I really feel like it’s yeah two sides of the same coin as we say um and there’s groups so Thomas iur at uh Stanford he’s actually working with with his students on these topics uh again very grounded

In formalism and Theory well that’s from the logic side of things of course and causality wouldn’t be much different though um but yeah that’s that’s also a bit missing actually because uh yeah uh I guess there’s a lot as you can tell you said about abstractions can you tell

Our audience a little bit more about abstractions why would they be important in causal reasoning and where would be the place where we could find them useful today so to me as a computer Scientist by training abstraction is like a key concept if I think of abstractions in a

Causal sense so if I think of this original paper in 2017 by Rin atal at uaii um it was not even called abstractions it was called like I mean in the work I think they Ed the word abstraction but the title wasn’t abstraction it was something like the causal consistency of structural causal

Models right so you you look at two different models two different structural CA models which is the centerpiece of pears formalism and you try to equate those under which conditions can you can you make them equal but in a sense that you have different levels of abstractions what

Does that mean so you have lowlevel and high high level variables they give this very nice example of something like uh cholesterol in your body right the total cholesterol that will be a high level variable and actually if we look more closely into the biology then you know

People know that there are these lipoproteins right HDL and LDL and they are kind of composing this total cholesterol and actually only one of them in a sense is bad for you and so these would be low-level variables and now you try to equate them to high level

Variables and the tricky part about the Cal stuff is that because in pears formalism it’s Interventional list right so you have interventions and um you need to make sure that you know you don’t just equate the variables correspondingly in this case LDL hdl2 uh as a sum to to to the total cholesterol

But also that you respect interventions which are possible right um and that’s abstractions in a causal sense as they defin it and then you know s be has generalized from there with his co-authors and so on and so forth um and just thinking maybe on an intuitive level broadly speaking well we

Always talk about you know graphs graphs and and stuff like this graphs consist of you know variables that are somehow connected to each other but where do like we always ask where do the graphs come from and I I have a counter question like where do the variables

Come from right what is even a causal variable I remember one researcher from Amsterdam Tako Cohen very famous also I’d say in the uh geometric deep learning uh community and and he is also doing a uh more research in celti nowadays and and he was also asking uh

On Twitter actually like what is the causal variable to begin with right the concept somehow didn’t make sense and I think this is where abstractions and maybe also just the the the whole question Paradigm of representation learning or representation come comes into play right like that’s

Why it’s important we we we are able to to look at things at a different scope um and and and and still capture their characteristics and and and that’s why I think abstractions are important because that’s exactly it’s a topic to study of exactly those things and it’s the first

Step essentially right so if if on an for an autonomous system right like we also make sense of some kind of abstractions when we interact with things we say oh this is this is a car oh this is a whole set of car which forms the congestion as we were talking

In the beginning of the episode right so yeah yeah so so we can have different levels of of description h i remember that in Pearl’s book causality not the book of why but the bigger book the town of causality as I call it in in in his

Book he he has this section I think um which which is a transcript of of one of his talks and he discusses the Russell’s B beron Russel’s argument against causality and he uh and he talks about this idea that causality could be a a convenient shortcut do you feel that

This idea of abstractions and those questions regarding causal variables are related to this idea of causality as a as a shortcut can you maybe just elaborate once more on a convenient shortcut with respect to what so there’s a rless argument against causality and um in this context in this context Perl

Says that hey maybe because it we don’t causality in in a sense like on the fundamental physical level if I remember correctly so don’t hey don’t shoot me if I’m missing it was a little bit uh some time ago when I read it but he says

That maybe we can Al maybe we can think about causality as a useful shortcut as we do in physics sometimes right so we have some reasoning in physics and we say like hey this is just a useful form about thinking about reality um that that brings us value so in this

Sense I I I I felt like when you talk about causal variables and levels of abstractions it doesn’t make sense uh to talk about a single screw in your car in any sense right if we talk about the property of of of traffic in in a large city because this is an emerging

Complex system at least we could think of it this way um so so that was my that was that was the connection that I had in my mind I was wondering if this is something that also Rings a above for you thanks so so now I understand the point

Uh so there’s two things which come to mind so first of all I think what was already implicit in what you just said um you know you don’t care about the loose screw you know a single loose screw in your whole car or if we look at

A larger system um what is hidden there is the thing that it’s always with respect to some thing right whether it’s measurable or not but it’s with respect to some some aspect uh the second point was actually um with an argument by by B chov and others right so yonas P

Dominque yans B chov they had this book on elements of caal inference MIT press 2017 and um they so banard for instance and Dominic they are physicist by training actually just like UDA although UDA I guess went more into the computer science Direction eventually um and and

And they have this table in the book actually where they talk about differential equations right and kind of the specifics of the physics of the system as being like the most fine grained and most containing um um level of abstraction essentially of um how we can you know

Use formalism to capture things about reality and and causality was the second rung on this one so it was what they would call a useful abstraction right so it was kicking out all the unnecessary detail let’s say I always compare this to modelbased and model free reinforcement learning so there’s this

Prime example of you know there’s a baseball player hitting a ball going at high speeds with a bat um that person is not calculating on the Fly you know probably um how they are going to hit at which angle and what not right it’s it’s on an intuitive level right and so

That’s the argument for model-free reinforcement learning that you don’t have to have a model right you just need to know what to do in The Next Step what’s the best action right and so in that sense if you then you know go into the second run where Keli is then placed

You kick out all these details of the model which are not necessary but still are sufficient to and necessary to actually answer about your hypothesis right actually also compare this always to to to Neuroscience so so my little s in Neuroscience was in a subfield called conomics where they are concerned with

Actually building a map of the brain right the conone and it’s a huge Endeavor it’s on the neuron level so you get all the you know the the somata the X aons the the the and that’s very detailed yeah the D rites and everything

Right so it’s it’s like on the on the on the micrometer scale right nanometer scale actually um and and and the kind of Base idea as far as I understood always was that um it might be too detailed but it’s certainly sufficient to to to capture everything uh about the

Brain’s functioning right and and that was always the idea there right and and so again it’s about like your hypothesis space and how you want to cut it down and so pet at Al also make the argument that Cazal is a useful abstraction and so I

Think they agree with pearl and I would also agree with pearl so physics might still hold other interesting abstraction levels and so on but then again that’s that’s the choice right the choice you make and let’s see eventually if we go down the road we’ll find out if we are

Successful or not yeah that’s very interesting uh you mentioned reinforcement learning and in particular model three enforcement learning where we maybe don’t care about having a model that that much but we are interested in understanding what is the next best thing to do and this brings uh to my

Mind the idea or the discussion that is all over the place happening all over the place now um about large language models what’s your position on this where do you find yourself in the llm debate especially from the causal point of view yes so on on a general topic

Right um also if I reflect on on what my adviser chrisan casting would say is that um maybe we should start you know being more transparent about this and actually giving our thoughts even in the papers right so I mean certain conferences do that similar things already so for example NPS right is

Actually doing you know these ethical statements societal statements and so on and so forth which you should include in the papers and I guess for most of the papers uh nothing is happening there of course if it’s llms then surely that’s a big topic um and maybe we should also

Start talking about like our philosophical grounding right like where do we come from as as as as persons right as as the people the human beings behind the science right just in a brief like self- summarizing way right so that we can understand like uh am I a

Supporter of the scaling hypothesis with llms right or not right so um I think this would be useful I mean I’m always against you know labeling people into boxes and stuff like this I try to get out of boxes myself um but I think that will be super useful actually uh this

Was on the general topic for the llm topic I’d say that um placing myself personally as um an advocate of llms I I like what they are doing as I mentioned earlier um but then again um from a causal perspective right I’m I’m really just trying to understand are they

Caused or not right so what we have seen certainly um I mean many of you have experienced it yourself when you’re using something like chat GPT or GPT 4 right um but also we in in in in our uh setting of of evaluating these models empirically um trying to be clever about

The ways we formulate you know queries that could you know have some kind of causal implications or give some causal insights we we find that you know sometimes they actually performing pretty well right so actually gpt3 and and and and other models were performing

Not so good but then you go to gp4 and suddenly well it’s performing well right might be that they already used the prior version of our paper in their training data now you know things can happen um but even without that even if it’s just the Improvement part right

It’s actually incredible and then you’re like okay so so when it gets it right why does it get it right and so that’s the question we were asking and that’s what we were investigating so in your paper you proposed this interesting formalism that you called meta SCM you mentioned three different languages

Formal languages that can be used to describe three letters of three ranks of the letter of of causation um that judia Pearl has proposed in his original work what are your thoughts about this formalism or maybe let’s let’s let’s take a step back back can you tell our

Audience a little bit about how to understand those meta scms and what is it about what’s what is this idea about so so I think the key aspect to understand and it’s a lot easier to understand first than meta ACM is this correlations of causal facts that’s the conjecture we propose right conjecture

Because again it’s a hypothesis that we believe to all true although we don’t have proof right the best we could do so far was empirics right and you know proposing a theoretical grounding in pear form ISM which would explain this right um but no definite proof so so

This is open an open problem um and this idea really stemmed from some intuition also Christian had um and then you know Mor and myself we picked up on this was that well let’s suppose they only learn correlations right then for them to answer some causal questions first

Correctly that would imply that there was some correlations on these causal questions questions and CA answers right and then again if we think about it a little bit on on this intuitive level a bit further following our our nose in German we would say Aisha then um it’s something along the

Lines of um you know I asked the question does uh altitude cause temperature in the sense of if I go up a mountain does it maybe become cold and the answer is yes right and then if you have a textbook which is talking about the physics of of these me mechanisms

Right then you can sure bet that you know if there is a question formulated that the corresponding correct causal answer at least in our world yeah there might be different worlds where you know the cause direction is reversed um holds true right so and and and and and following this intuition essentially we

Just state that well llms are actually training on on in such cause and knowledge right um and and and just by thinking about it from a different perspective it seems to make a lot of sense intuitively because we do experiments right we find out this mechanism that you know the the

Molecules start moving more slowly as you go up and so on and so forth that um yeah there is a link from altitude to temperature and then we write this down right this is like how knowledge is being passed on from generation to generation we have vikipedi it’s an

Encyclopedia right for for all kinds of different knowledge and actually we we found this example this very example we have it in our paper um about lapse rate and and these kind of physical Concepts uh you have articles on them so the knowledge is there right and if we learn

To predict the best next word well if you know I’m talking about you know altitude and temperature um then causing is probably the better word than not causing right or or formulation for that matter and so this is where this intuition comes from right and and also

Then with now our formalism of the meta to be a bit more technical but for the technical details please go into the document itself we link the paper in the show notes um and so there the idea is uh well it’s it’s this it’s this idea but like conceptualized and formalized a

Little bit differently so what we are saying is Well we have these rungs right and rung two and three are considered causal in in in the sense of Pur so there will be interventions and counterfactuals correct and so sem though is a very general form formalism right even if we you know or

Or especially if we look at work by for example bongas and Peters and others who have you know looked at cyclic causal models and then they have really captured what what we mean by structural causal model right and it’s still even in their formalism with the measur of

Spaces and so on and so forth it’s very general and so again we you know as humans we do experiments we get some causal insight and then you know save it in some taxal representation and while you know we can have this debate of understanding versus knowing right which

Is a whole philosophical thing you know related to the Chinese room argument by John Sal and so on and so forth um it’s already this aspect of um not that that we have this knowledge safe there as a representation and well this could also just be a

Variable in our SCM now right our SCM has a variable whose domain is all these statements right and or just texture representation and specifically it’s these statements right and so we just conjecture that essentially llms are training on correlations you can find in the training data of these causal facts

And so now if you look from an sc perspective from a formal perspective it’s essentially always a dance between two models one is the regular SC right which is something like you know altitude temperature and so on and so forth and I find out that there is a

Link from variable a which represents altitude to variable T which represents temperature right um again an SCM is a more General formalism it implies a graph structure I’m talking about the graph now for the moment and now imagine I have a second second model and this is

What we call this meta SCM which is kind of a hierarchy level above right which is um talking about this insight about this whole graph itself about the Assumption right that altitude that a causes T yeah MH and and that’s why we call it meta right because it’s an SCM

Which is on a meta level on a level higher than regular scms talking about other scms so if you depict it graphically in a sense uh you could say that you know you have these different ranks for the SCS but one is just shifted above right and so the L1 of

This one is connected to the L2 of the other one mhm I was curious did your your let’s say your prior regarding the the conjecture of meta scms has it shifted after after gpt3 was was released no I’d say no no so if I elaborate a little bit on this

So yeah I think my personal belief right um about I guess large language models I’m certainly biased by you know I guess the nature of causality being a symbolic thing um but also then you know the perspectives of my adviser and my colleagues MH um so I’d say still like

Personally if I ever have to take it I I would say like okay for me it’s already this conceptual difference right between the texal representation of knowledge and the data and the experimental side of things right so again we also have this example it’s like this intuition this nonformal part

Of the argument in our paper because well it’s part-time philosophy and so you know a physicist has their setup right like we have the setup here with the microphone and camera and everything and we record stuff we measure stuff and then you know we look at the data we

Think about it and we conclude you know something that comes out of it right it’s like the symbolic regression as we call it in AI that we are doing um but then we write it down in textbooks right like we have one beautiful book here as well yeah and and and that’s now

Knowledge that you know if I trust the source and assume it’s correct essentially and let’s say for the matter it’s actually correct right then well if I learn this fact now and someone asked me a question and a test whether I learned it myself by doing the

Experiments as the author did or whether just learned it from the author doesn’t matter anymore right so the be behavioral aspect is not in the it’s indistin shable right it it it collapses to to to one point essentially um and so this is complex because we also have the

The entire societal structure around this right to say that I trust this auor there’s a number of conditions that need to be met and we have heuristics to to reassess our uh our trust in in somebody’s uh statements and so on and so on right so exactly and that’s why

It’s a philosophical topic I think it’s a deep philosophical topic um but as we see uh even Ai and llms are not unfaced by philosophy yeah definitely um so what’s what are your thoughts about scaling lows so so we we can see there are some papers from Microsoft research teams uh

For instance two teams uh actually published papers on on this showing that uh those larger models in particular GPT 4 uh can do pretty good job in in causal I don’t know if you can call it reasoning but they can answer causal queries in a way that seems relevant to

Us and and much more so than the previous generations generations of those models so one obvious path is to think about the size of the training data and how many of the testing procedures that’s also your hypothesis in the paper uh could have been included

In the training data of gp4 but it seems that also gp4 is just making just doing better job in many different tasks that require some form of world model or at least we could think that they require some form of of the world world model what are your thoughts on on this I

Think I want to kick this off with uh just um talking shortly about one aspect which is important to me um from emra and amids and and the others work so in in their work right they they looked at llms and and their causal Powers right and essentially

Concluded also something like that on the tubing and data set uh you know they perform very well um and I’m very critical of this right so I’m um I’m positive about the the fact that yeah sure I mean I I believe these results I think they’re republe and this is

Scientific results but I don’t agree with the implications or at least you know the way I’ve perceived them so you mean you mean the conclusions that were proposed in the paper or correct so essentially I had the impression at least also from my Twitter interactions with Amit that

Um we have seen some high relatively High uh accuracy on on this data set and so now we kind of uh say like wow this is incredible MH um but actually if you dig down and and look into this data set so the typic pair data set I don’t know

It’s like a bit more than 100 I think uh different pairs of XY variables so it’s a b variate data set the task is to conclude whether X cause y or the other way around and you know while there’s you know examples which are very clear uh something like altitude and

Temperature for instance it’s actually in that data set and and there’s like also the the do vadin data which is you know like for different places temperature and and their corresponding altitude recorded but then there’s very obscure pairs mhm so I recently gave a talk actually at the Paris workshop on

Cality mhm uh cality in practice and um there I was mentioning this so I have it on my slides so I don’t know the details right now of which one it is right but it was very obscure so so there was something like um I don’t know the

Something produced by some you know object X at at time y that was just one variable right and then the other variable I don’t even know what the concept was and and and and then surely for for such thing um especially say there’s names in there right like like

Just like in the famous Basin Network repository there’s this Basin Network where there’s like the earthquake diagram where there’s like whether you know John or Mary would you know uh call the the the firefighters if there was an earthquake right well the llm does not

Know who JN is right so as soon as you have this I think this invalidated completely right but in general these Concepts were just so obscure that um well it’s it’s just a guess and it’s a binary guess right so either X Y or YX right and then again we also just

Looking at accuracy right and so if you look a bit more closely in the details you start to realize that um well this doesn’t mean anything the the output that I get now right sure it’s it’s it’s cool to see that it works quite well but

I think that’s the only only point that I want to emphasize on right that um we should not base it on on on such not just like like very simple metrics like accuracy but especially like we have to consider what data we are basing it on right this

Is uh for me always an important concern just like someone who’s in applied machine learning uh would tell you that uh working with the data pre-processing cleaning and and and things like that right uh and usually the the biggest challenge right of course with deep learning is the Hope right that you know

We have so much data that you know it just mitigates itself still biases will exist but that’s you know a human problem actually so uh yeah to to conclude to that question that’s what that was my only only point to that one yeah and and regarding the

The other tasks that we used in those in those papers so there was like this counterfactual reasoning Benchmark and so on and so on would you would you say that this kind of argument will also apply applies there uh no I was just specific about the tubing about the TU

Exactly correct so in this in this context and in the context of your own work and also the uh the updates to your to your recent paper uh what do you think about scaling loss going back to the original question yes yes so they are kind of part of this

Whole discussion right so if you’re like a connectionist or or in general like in neuroscience and you know functional connections and things like that you know on on average there’s like a 100 billion neurons in our brain that’s a lot right um and each of these and this

Is now actually the the most impressive part because actually if it goes just by cell number you know your your liver would have more cells actually right and and also some other mammals would have more neurons right but we consider ourselves I guess more intelligent so um

It’s not just the number but the number is definitely Cru crucial factor it’s actually the connectivity so on average each of those neurons has like a thousand connections to other neurons and it’s a big social network and uh that’s kind of the grounding also for because these are neuroscientific

Results right and this grounding for you know something like the scaling hypothesis that essentially the models if we put it into relation they are not quite there yet and scale is all we need right uh there’s this funny meme also which uh uses also the bitter Lesson by Rich

Sutton who is one of the uh Pioneers in reinforcement learning and it just goes like GPU go and it should kind of depict that you know it’s it’s just going like yeah gpus and and and put them to the test get the temperature up and overheating and then and then you’ll get

There um my personal take is um I guess also biased just by the Coeli side of things that um with the symbolic part of of the equation um which means that uh I believe that we still need Ingenuity and uh conceptual development and that it’s going to be combination of both and that

Scale is certainly necessary I mean we as humans are certainly an example of that um and also if we look at something like the neocortex right like the outer layer the human brain and then how it’s just like even twisted and that just gives you more highways essentially interesting connections that you can

Form right so connectivity is super important and we have seen so much success already with deep learning so so why not just push it further um but in my humble opinion it’s definitely a combination of the both right just what like neuros symbolic is trying to do and

Given that I see the connections to the logic Parts I was saying like neuro causal and symbolic in your paper you also mention causal models in the context of Black Box black boxiness of of of uh contemporary neural networks we often talk about people that people are good at causal reasoning I’m

A little bit skeptical about this uh but this is a view that appears here and there in the community from the psychological and and neuroscientific point of view we have pretty good evidence that humans are sometimes black box reasonless for themselves as well do you think that

Being white box is necessary for causal models and and if if so why would that be very interesting question deep question so I’ll have to think for a moment so maybe I’ll start with the human aspect so I think I agree that um humans might not always be as good as

They might think themselves to to be in causal reasoning um or in general in causal reasoning so of course naturally we are good in something like a personal experience so so my bike got stolen when I was still studying um it was partly my fault because I know that the chances

Rise when when it’s night time yeah um and I still still left it all the night I stayed at University the whole day um and then when I arrived it was a rainy day and my bike was nowhere to found to be found um but yeah in that moment what

I say is a damn it right if if if I had come sooner my bike would still be here that’s the counter of factual right that’s also the examples you’ll find in in the book of why for instance so in that sense we are very very good right

But if we think of policy and more complicated topics right then it does not feel that agents human agents act causely right um if we go back to the to the to the other point right so um could you just repeat maybe once more like the the

The key aspect you you you want to discuss here yes so I was um I was thinking about humans as not necessarily white box reasoners in in general and in causal sense as well and maybe to expand also a little bit on what you mentioned there was a there was there were many

Interesting experiments in this area but one that I had in mind when I thought about this question was one by uh Michael gatena who was um experimenting with patients who had uh whose um hemispheres were disconnected and so these patients were sometimes primed just one eye

Because eyes as you know but maybe some of the listeners are not aware of this um we have this cross reference in the brain so left eye goes to the right hemisphere right eye goes to the left hemisphere and so on and if we destroy the connection between those hemispheres

Which is um which connected by a a part called Corpus colossum then those hemispheres are largely independent and so the researchers were were priming the the participants of the experiment by showing them just to one of their eyes uh some object and then they were asking them to um say as tell

A story or maybe I don’t know asking them for a reason why something happened and they were making up very plausible uh very plausible explanations causal explanations why something happened why the true reason apparently was just because they were primed but they were not aware of this so I don’t want to go

Into neuroscientific details like why they were not aware and so on and so on but that’s the that’s basically basically the case so this means that we might be very good in coming up with causal explanations but they are not necessarily they might not be necessarily relevant and and uh

Sometimes we might have also good explanations and might we might be not aware where they are coming from and they might be relevant um so my question was given all of this and given that we as a machine learning or AI Community we are looking

Up to to how humans uh function in all different different areas how do you think uh necessary is the the white boxes versus black boxes box to to causal machine learning to causal AI to be useful for us so I think it’s um almost implicit also in what you said

Now that uh White boxiness and explanations are kind of intertwined but they are still independent Concepts right so what we seem to do as a community is that we want to be white box because we want to be able to understand these systems right and we

Expect that from a white box system we can actually do that but but who tells us this is the case right like even if we just go from the cognitive science perspective right the study of the the human cognition right then which is again a twin discipline to AI right

Essentially in in cognitive science this twin science of AI um we are already asking a question what is an explanation right um it’s not defined right there’s definitions you can propose right but then you can find counter examples as often times in philosophy um and so we’re trying to capture the thing which

Does the best job in a sense right and so even if you were to have a white box model doesn’t mean it’s explainable right uh I think a very prominent example for this are linear programs so linear programs it’s kind of like space class of uh mathematical optimization

Problems um you have like a cost function constraints they all linear and we can solve these uh systems uh we have algorithms and and and also just the whole thing is white box in nature right uh it’s it’s not like we don’t know what’s happening but still so

It’s white box in a sense that we we are very well aware how the structure is how the algorithm is structurally constructed and what is happening and how the signal is being processed at each of the steps correct it’s completely explicit right but now and and even if I it’s very interpretable as

Well if I go for a simple example but now I scale it up actually this is one project we’ve been working on which is called Plex plane here in Germany funded by the government on um it’s concerned with the like uh uh energy goals of Germany you know to be

Kind of climate neutral by 2050 and so a lot of uh machine learning in also you know Energy Systems is based actually around LPS which are just very large and and now the motivation is there to understand you know why for example we need more photovoltaics then we need uh

Market BS uh electricity for example right and so here now you have a white box model still a white box model but it’s at a scale which humans just cannot process I always think of this quote by UDA Pearl which says like um you know I cannot even understand five variables

Let alone a thousand or more right so that’s why you often times in the example see only maybe up to five variables right and so explanation does not go along White boxiness we we tend to believe so and it’s certainly more accessible because of its explicit nature than blackboxes but doesn’t mean

It’s the case right and so that’s the hope and that’s why I just want to touch upon this point because I think it’s already independent and then again this whole aspect of explanations are not even defined right in in in in some kind of um ultimate sense right surely

There’s a lot of useful explanations but I think what I’m trying to get it is that uh maybe this is a hot take now that you know blackbox can just be fine right as as as long long as we you know get explanations from that system that

Are faithful and that do the job right and so actually we also have some Works in this Direction that’s why I’ve was coming also from from different perspectives and so mhm and what are your conclusions from your work so far so our work is uh we we proposed a a a

Recursive algorithm called structural causal explanations uh we were bold to put the structural causal in there why because um it’s it’s essentially an algorithm which uses the graph structure but also quantitative knowledge about you know cause and effect relations for example if it’s linear causal models then it’s just the coefficients right

And then you know you have a question say we are in a medical case right this the example we do in the paper as well so so we have different patients and record some kind of data about them right we capture them in different variable representations say for example

The age of a person this we can naturally represent as an integer number for example the then we have some knowledge about say their nutrition um in some kind of numerical sense right say a high value says that the person has a relatively good nutrition right it’s balanced by any dietary standards

Right we can measure maybe some kind of uh um key indicator of overall health and mobility and and and then you know you could have you know a set of patients and then say there’s a patient called Hans and now you uh see that you know

Hun is an elderly gentleman um and his health is overall not that good actually also mobility and stuff like that and now the doctor might ask the question well why is hans’s mobility so bad right and and that’s kind of a relative notion right because the doctor is comparing it

To some kind of standard right and that standard can just be like the average uh mobility of the whole group he’s considering or over overall what we consider in society yeah and then what this algorithm does is it traverses recursively through the parents essentially right and you know it would

Give you automatically then an answer which is something like you know hans’s Mobility is bad actually because of his bad health and then again this bad health because of being elderly right and mostly because of that although actually the food habits the the nutrition is good right so you have like

Both the the structural knowledge right about the parents but also then you know it’s causal because we grounded in a causal model that’s the modeling assumption right of course if that model is wrong then you know you’re conclusions are not sound but what we are proposing is again this causal

Inference part of things now we look back to the beginning right that the conclusions we make should be sound right so whether what we what we base it on is is true or false that’s a different story but the sound conclusions is what we care about and

And and then yeah you have like the traversal the structural you have the causal and you have also a little bit of quantitativeness right because you know food habits if if they are better then you know that usually improves Health right um if you’re old that usually uh

Decreases health and so on and so forth right so this algorithm also gives you information about So-Cal actual causation what what caused this outcome in this particular observation so this is a great point that you mentioned because this work is also still under review MH um and one of the reviewers

Was actually coming from this actual cation area and while we do have a comment in our work on this we never um explicitly go in details through this uh in anyhow I personally believe the paper already packed a lot so maybe we should split it up actually and the actual

Causation is not happening here right so so you’re talking about individuals but it’s not the actual causation formalism by halpen uh and and pearl or or or nowadays mostly halpen so informally you can you can speak about this but informally it’s not it’s not a formal so

You compare to the formal you can compare it right it’s definitely individualist causation right not type causation which is because it’s not talking about the population but it’s talking about individuals um which becomes very apparent in this particular example I was giving just now um but uh

It’s it’s not an actual causation work right it doesn’t satisfy now at least we haven’t checked whether it satisfies the actual cation axioms mhm you you mentioned this example of of of a patient and and patient being in some in some context uh their personal context

Of of the diet context of the medical record and so on this is an example that puts cality in a very very practical place what are your thoughts regarding adoption of causality we are definitely more advanced with adoption in Industry than we were even one or two years ago I

Think things are moving very fast but still there are some obstacles that people meet on on the causal Journeys or organizations meet on on the CLE Journeys what what are your thoughts on this so I’m a bit split on this actually so on the one hand I think there’s so

Much work left to be done right and I guess that’s the scientific perspective that’s the philosophical perspective as well right it it just feels to me that there’s so many unresolved things so although we have made tremendous progress right um and well that’s also good thing because well that’s my job

And and so there’s more things to do right you’re never running out of of things to do um so so in that sense I I feel like okay it’s not very practical right there’s what what feels at at times like an outcry by the community that we need benchmarks you know to to

Measure the whole you know machine learning AI thing of objectives right um nowadays I’m a bit contrary to that as well because I have read this book um the myth of the objective right uh which is talking about this idea of objectives actually being a false Compass when they

Are ambitious I’m not going to Define any of the terms right now um please check out the book it’s very nice book by by researchers who were based on in Florida I believe um so so so this whole objective thing it’s it’s still kind of like yeah and that’s what a benchmark

Does essentially and you know the the classical machine learning way um so but in general yeah I I agree you know gold standards are somehow missing but then again well that’s the whole point right like we did physical experiments we found out about some laws about other things and that’s our causal knowledge

Now our goal standard right like if we had this then we wouldn’t have a problem in the first place so that’s that’s my kind of like I guess Academia perspective on it right um the other perspective is actually that I think we are already super successful

Right I mean sure we we can never evaluate and again this is the objective b we can also never you know um uh we’ll probably never have like this ultimate truth kind of thing right that’s again a very philosophical debate right um relates to to works by by goodle by Kor

And and these tremendous mathematicians and logicians um so if you think about definite mathematical proofs that could be applied so that’s more from the AC academic perspective and and this one it’s more about um so I was trying to to get at the point that I think we are

Already doing very great and it’s not perfect by any means and and that this discussion of perfect and Truth is anyhow a different one which we might never resolve right I guess my personal take would be it’s like very likely and so that essentially if you look at practitioners in causal

Inference right like so for example I I remember going to Keno for for for cality workshop and and we had practitioners there who were applying it to you know again biome medical data and so on um and they had huge grafts like discovered with causal Discovery right

Um and and validated with experts and and things were surprising there right it reminds me of this one work by by Peter vitkovic and and and others at Deep Minds right where they looked at you know uh science X AI right like math X AI where essentially uh they didn’t

Use any fancy techniques per se right but they just applied them thoroughly and consistently and then always caught back to the mathematicians and in this way the mathematician could find some representation which was more suitable for making the next big step in you know a certain theorem right and that’s why

They got some new results right and I see it in that kind of way that it’s our assistant right so what UDA is also doing a lot later right like trying to do this personalized medicine and and these kind of things which is really the the yeah SC AI scientist or the AI

Assistant in a sense right I guess that’s also what a lot of AI researchers Envision in in the future um yeah you mentioned Peta vich kovich how was your work with Peta and what was the origin of um of your meeting and then your common joury your your your journey

Together yeah so Peta is a really nice guy um amazing lecturer um I remember watching some of his lectures uh on on you know geometric deep learning graph new networks um and then I was actually just a participant at the Eastern European machine learning summer school

By the way amazing summer school so if anyone wants to you know go there you should try apply uh I I was part there I was like once a participant and then the second time I was uh actually also part of of the lecturing and both times were

Blast from both perspectives um people are very nice and and you learn a lot and and and he was actually uh one of the mentors there but you know he was always also giving lectures and I just reach out like on the slack Channel it

Can be as easy as that right and uh and then yeah we we we got together I I told him his my intuition about you know how Graal networks and structural CA models have a relation and and that’s where we where we picked it up right and

Eventually wrote a paper it’s my most set of paper so far although it’s not published uh and yeah that that that is how how the game can go right um but we improve right we revise uh I’m sure eventually it’ll go in somewhere um but

Yeah that that’s how it end up and it’s pretty cool what were the results of this P this paper this work what that you that you have in with Peta something that you find the most important outcome of this uh this this project with him so

To loop back to one of the things I said earlier uh when you asked me the question of uh what are the most important things I consider now in my humble opinion about cality and how the scientific research is going to continue I was talking about you know thinking

Outside the boxes asking the questions that have not been asked before right Making Connections Building Bridges not burning them um and so I think this is the most important outcome of this work why because we brought a bridge between one of the hottest fields in deep learning geometric deep learning right

Graph networks and things to causality to structureal cause and models right like we coming rather from a causal perspective but then again don’t get it twisted we are also heavy on the machine learning it’s the artificial intelligence machine learning laboratory after all and that’s that’s my original training

And yeah we we we we we Tred to find consistent way of connecting these two you know uh Concepts ideas Frameworks um and I think we succeeded and this would be the most important contribution that essentially you have a bridge now between these two Fields you can you can talk about these

Topics you open a whole new uh research Direction M we’ll link to the paper in the show notes as well so everyone can read the paper themselves and you mentioned that graph Neal networks are are one of the hottest sub areas of machine learning today in your paper on

On caal parot you s CH and the quote is early dramatic success followed by sudden unexpected difficulties this is a description of the typical life cycle in in in machine learning research um what do you think about what do you think and how do you feel about those hype cycles

That we that we have in machine learning do you think this is something that is useful uh or maybe we would be better off if we just try to temper down emotions a little bit so I think what my advisor Christian would say now and I

Guess I would agree is like ride the wave and and just push for it um I don’t think it’s stopping this time so actually there was some study which uh suggested there’s like 30 years cycle always when the next AI winter as we call it historically speaking uh occurs

It seems like a COR correlational measure and so this if we look at the a history right it started arguably back then with John McCarthy uh Minsky and all these other Pioneers who got together at dot mouth for this like summer project in the 1950s um they were thinking okay if we

Put a bunch of smart people also including people like Claud Shannon right like I mean information Theory right entropy and all these kinds of Concepts in in this sense that we could solve AI right and well this didn’t work out but at least they established the field in a sense right that’s closely

Connected to historically speaking also just to to computer science and its Origins right um and then you know there’s always this right the perceptual you know neur Nets and then you know again it doesn’t work and there’s a winter and now we have the symbolic but oh it’s inflexible it

Doesn’t work winter right and so on and so forth but nowadays it feels at least I in my opinion right I have not lived in these times right I can only report from reports that I’ve read um and accounts that I’ve uh considered but it doesn’t feel like it’s it’s slowing down it’s

Just increasing and going to the buzz aspect of it that you were mentioning with all the Heat and everything um difficult topic so so on the one hand you can make the case then well we need this right then doesn’t matter uh there’s no bad publicity right like in a

Attention kind of sense but then again you can say okay well we we try to be factually correct right we we endorse being reproducible and and have certain scientific values so so we should tone it down um actually I was just told by by a colleague who was doing a PhD in

Tubing and that uh their AI or machine learning department is is is is actually trying to decouple itself a little bit from AI because of all these implications because they want to be on the defensive I personally think this is wrong why because it should be discourse

It should be you know scientific debate and you can still not make these statements and and separate yourself from these statements but now to start separating AI from ml I think that’s not possible and should also not be done you mentioned about uh putting many smart

People together in one place in in a hope that they will solve a problem so this is also uh somehow related to the to the idea presented in the in the Oppenheimer the movie have you seen the movie yes I’ve seen the movie uh I considered more of a documentary yeah I

Was aware of of the history of the characters uh without spoiling now because again I see it more as a documentary these things happened right um I find it nice when when we saw the scene at the Institute of advanced study in Princeton where Einstein was actually

Having a walk with cot girle right um that’s that’s a little detail which really showed me that Chris Nolan and his team did an amazing job of of portraying this movie so so yeah I watched it and I liked it then the three

Hours went by in in a in a rush M do you have any personal Heroes among the people the the character that were poate in the movie so so I think I should be taking a hot take now because that’s the thing right with Heroes right like we

Trying to idolize certain people and so on and so forth um and there has it cons and and and Pros right as with everything um but if I just take my personal personal state of it’s difficult right I have not known these people right I’m I’m judging based of of

Just what I’ve read and and and seen and of course the movie you know has these movie aspects to them of being more dramatic and and presenting in a certain way um but I think really yeah actually the person I mentioned just now K girle one of the Godfathers of uh Lo logic

Yeah with his incompleteness theorems following the uh David David Hilbert program right um tremendous results in incredible what a life story actually I’ve read a biography on on his right to um I think the book was what is it called the the the journey to the the

Edge of Reason or something right even the title was just like wow um and actually there’s a lot of funny stories about him so so we were talking about naftal earlier so while I was going to to Munich for the the cality for ethics and Society uh Workshop uh which was

Co-organized by naali Weinberger um I was uh reminded by by a person David uh talking also about goodle that you know when he went to to the IAS eventually in the US uh when he had to take this citizenship test that he actually found a flaw in the

Constitution in the US Constitution and it was and actually this was mentioned in the book but I read the book a while back so so I didn’t remember the details but he reminded me and it was essentially that the definition of of of a year was somehow not clear so so you

Could redefine it and then actually you could have the government uh well uh rule the people for this year which would now be well practically infinite right and so and then you know people like Einstein and the others right had to hold them back don’t mention this

During your test because you just want to pass right you don’t want to freak out the people so um yeah I think from that movie particularly I think goodle would be my hero yeah I think I mean his his his his his passing then was was

Quite sad um you know he was it it seems depriving himself from food right because of of worry of poison and and these things and it’s a reality I mean for all these individual if you just look at the human side of things that’s very very important to me as a as a

Researcher as a person right um it’s a har story but the achievements maybe especially in light of that are just incredible so if I had pick I’ll put my money on C what are what are two books that were most influential for you for you as a

Person it can be during your career but also your personal development definitely what I’m picking is the book of why because it got me sitting here right now um and I really I it sparked the fire in me it really ignited I I think whatever UDA had in mind he

Definitely achieved it on that day with me um because I think what he had in mind was you know inspiring people to to study the science of cause and effect for the second book there’s so many books so I I I like reading lately I’ve

I’ve I’ve not read much but in the past I I would always have my Kindle um because it’s easier to carry I also like physical books of course I like them actually more but then again you know you have backlight and stuff like this so it’s pretty good um for the second

Book it it would be hard to choose I mean I I love a book like from from Monroe on on like what if right like the xkcd comics and you have like these amazing contrasts right still I also love a book like thinking fast and slow

Which got cited I believe way too much and and and also in inadequately at times by by scientists in AI right right now um and and and the in the recent past um but then again I’m also thinking of of of of things from G you know we

Actually sitting here at the University campus of of G University and uh Johan W from G I mean tremendous poet thinker in in general right like it really reflects this I guess this historic German um traditions on on and values for for for science for for discourse

For philosophy um and you know A Book Like fa right I mean we had to read this during school right but it was actually a fun read because there’s so much hidden in these books actually one of his lesser known books is the besta Divan which is like the the West Eastern

Divan and danan in the Arabic Islamic sense um it’s actually he he read a translation by hais uh Persian uh you know uh poet God essentially yeah um and and and and and and works of of half is were were were translated by by an noing

Guy which then fell into the hands of GTA and and he felt like this was his uh twin his twin brother in in in the mind and so so so you see G referencing Islamic things and it’s just a beautiful combination of of you know like Multicultural thing right and and

Historically and so maybe it’s even a book like that that could be my second favorite but for first I definitely choose the book of why when you think about people who are just to causality what resources would you recommend to them in the beginning how to start so i’

Say you go to discuss. quality. link that’s uh the link to the landing page of the cality discussion group and actually at the bottom of the page there’s like a big section a list with all kinds of resources but just to briefly for the sake of the podcast

Reiterate on this um essentially it’s the books right so the causality textbook standard book by Pearl um it’s a big big and dense right so I more use it as a reference book and and read here and there I would not go in like one

Goal I mean sure go ahead uh if you’re coming more from a machine learning perspective I’d propose the elements of causal inference right by by petal um amazing examples uh really like uh I enjoyed the book a lot um other than that there’s a very nice survey by a

Friend of mine Jean kadua and and his colleagues uh which is particular for machine learning right but still it’s uh again if you care about this then it’s a great way to actually get to know papers in this area right like tremendous Works uh Works which were influential and and

Popular um and then I guess there’s some lectures I love the lectures by Yas pet in general as well he’s a really amazing lecturer um but also by elas bar Bo um so yeah check out these references and then also just use the select channel so you can just ask for anything particular

Right or reach out to me or or Alex so yeah some people come to causality uh and they are really really passionate about this topic but maybe they feel a little bit discouraged by the way that sometimes causality is fought in a very formalized way what would you say to the

People what would be your advice uh for them to move to move forward that’s a bit tricky one because while I would love to say that you know don’t worry you don’t need it I think you actually need it so you know I think UDA Pearl would you need it you

Need the formalism right for cality because I think how UDA Pearl would phrase it himself is that it’s a language it’s a language a formal language a mathematical language with mathematical notation to talk about modeling assumptions you know about of the data generated process and and and and and finding you know um

Expressions and and sound conclusions right thereof so in that sense formalism seems necessary and and then again I guess well it’s the standard that we have uh in the scientific community in AI nowadays um and since it’s also more foundational grounds ground research closer to the maths closer to

Probability Theory right so we have measure Theory and then a special case probability Theory and now Within probability Theory we have causality and actually there’s been a recent work by by Jun and his colleagues from from tubigan who actually did an aati uh of causality but not the pean

Framework actually right so they do interventions and these things but it’s a different notion um properly in probability Theory right so so satisfying even all the uh the precise and and and and the pure maths MH so I think you don’t get away from the fism but just talking

About intuitively I think you can get a lot from there but then it’s a double-edged sord right because there’s a lot of wrong intuition I believe as well that we have about causation mhm as famous examples in philosophy would also show is there anything you would like to

Say to the community of of people who are I don’t know just starting or maybe they have have started a little bit a little bit ago um how would you encourage them to continue their their Journey thanks for this wonderful question because I think this as we call

It in German the Appel is an amazing uh thing um so I’ll speak to the camera because I’ll just treat it as the audience right now um if you’re passionate about this if you think this is Meaningful this would be fun I think

It will be fun so do it just do it as Nike would say mate it was a pleasure thanks for having me it was a lot of fun yeah until the next time I guess definitely thank you for staying with us till the end and see

You in the next episode of the caal bandits podcast who should we interview next let us know in the comments below or email us at hello at caal python. stay caal